www.nature.

com/scientificreports

OPEN Application of machine learning in

breast cancer survival prediction
using a multimethod approach
Seyedeh Zahra Hamedi1, Hassan Emami1, Maryam Khayamzadeh2, Reza Rabiei1,
Mehrad Aria2, Majid Akrami3 & Vahid Zangouri3
Breast cancer is one of the most prevalent cancers with an increasing trend in both incidence and
mortality rates in Iran. Survival analysis is a pivotal measure in setting appropriate care plans. To the
best of our knowledge, this study is pioneering in Iran, introducing a multi-method approach using
a Deep Neural Network (DNN) and 11 conventional machine learning (ML) methods to predict the
5 year survival of women with breast cancer. Supplying data from two centers comprising a total of
2644 records and incorporating external validation further distinguishes the study. Thirty-four features
were selected based on a literature review and common variables in both datasets. Feature selection
was also performed using a p value criterion (< 0.05) and a survey involving oncologists. A total of 108
models were trained. According to external validation, the DNN model trained with the Shiraz dataset,
considering all features, exhibited the highest accuracy (85.56%). While the DNN model showed
superior accuracy in external validation, it did not consistently achieve the highest performance across
all evaluation metrics. Notably, models trained with the Shiraz dataset outperformed those trained
with the Tehran dataset, possibly due to the lower number of missing values in the Shiraz dataset.

Keywords Breast cancer, Survival prediction, Deep neural network, Machine learning

Abbreviations
ANN Artificial neural network
BC Breast cancer
CNN Convolutional neural network
DL Deep learning
DNN Deep neural network
DT Decision tree
ER Estrogen receptor
GBM Gradient boosting machine
GBoost Gradient boosting
GRU Gated recurrent unit
HER2 Human epidermal growth factor receptor-2
KNN K nearest neighbor
LDA Linear discriminant analysis
LGBM Light gradient boosting machine
LR Logistic regression
LSTM Long short-term memory
LVI Lymphovascular invasion
ML Machine learning
MLP Multilayer perceptron
NB Naive Bayes
PNI Perineural invasion
PR Progesterone receptor
RF Random forest
RIPPER Repeated incremental pruning to produce error reduction

1Department of Health Information Technology and Management, School of Allied Medical Sciences, Shahid
Beheshti University of Medical Sciences, Tehran, Iran. 2Cancer Research Center, Shahid Beheshti University of
Medical Sciences, Tehran, Iran. 3Breast Diseases Research Center, Shiraz University of Medical Sciences, Shiraz,
Fars, Iran. email: [email protected]; [email protected]

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 1

www.nature.com/scientificreports/

ROC Receiver operator characteristic

SVM Support vector machine
XGB Extreme gradient boosting

Breast cancer remains a leading cause of mortality and morbidity in women globally, accounting for approximately
24.5% of all cancer diagnoses and 15.5% of cancer-related deaths1,2. Notably, 2020 marked it as the most prevalent
and lethal cancer for women in various countries3–5. Despite an unwavering rise in incidence, mortality rates
have fortunately stagnated or declined in recent years, potentially due to advancements in treatment modalities
and the widespread adoption of mammography screening programs, particularly in developed nations6–8. This
paradoxical landscape of increasing incidence alongside stable or decreasing mortality underscores the critical
need for accurate prognostic models, particularly those capable of predicting 5 year survival.
Despite a rising breast cancer incidence in Iran, underprivileged provinces experience slower increases,
plausibly due to limited diagnostic infrastructure. Paradoxically, mortality rates currently remain lower in these
regions. However, recent data suggest a potential trend reversal, foreshadowing a future rise in mortality within
these communities9,10. Furthermore, Iran exhibits a younger age of diagnosis compared to many developed
nations by approximately a decade11. Five-year and ten-year survival rates are estimated at 80% and 69%,
respectively12. These discrepancies in survival across regions likely stem from disparities in early detection
initiatives and access to adequate healthcare facilities13,14. This complex landscape underscores the need for
tailored predictive models that account for such socio-economic and infrastructural variations.
Accurate 5 year survival prediction remains a critical, yet formidable, challenge for oncologists15–19. This
task lies at the heart of personalized medicine, informing crucial treatment decisions impacting medication
selection and dosage regimens20–22. Breast cancer prognosis remains a complex tapestry woven from diverse
factors, encompassing patient demographics, tumor characteristics, biomarker profiles, and lifestyle habits7,23,24.
Machine learning (ML) and its subfield, deep learning (DL), which involves algorithms that analyze data
in a manner like human reasoning25, have garnered substantial traction in oncology, particularly in the realm
of diagnosis and detection using image processing26–28, and survival prediction29–32. These technologies offer
compelling advantages, potentially aiding healthcare professionals at various treatment stages33–36. Notably,
they hold the promise of enhancing technical parameters (e.g., treatment quality and speed) while generating
valuable clinical insights37,38. Accurate survival models empower physicians to streamline decision-making,
potentially minimizing false positives/negatives. For patients with lower predicted survival, this could inform
the consideration of less invasive treatments with reduced side effects39–41. To our knowledge, there have been
limited studies addressing both conventional machine learning approaches and deep learning methods for
predicting breast cancer survival using non-image data. Building upon existing research (Table 1), this study
aimed to develop and compare the DL and ML models for predicting 5 year breast cancer survival.

Materials and methods

This section represents the characteristics of leveraged datasets and outlines the step-by-step procedures
employed, encompassing the entire process from dataset preparation to model development and evaluation.

Data source and dataset characteristics

The dataset in this study included the data of 2644 patients from two centers. One of these centers was the
Breast Diseases Research Center of Shiraz University of Medical Sciences, which supplied data of 1465 patients
from 2003 to 2013. Another center was the Cancer Research Center of Shahid Beheshti University of Medical
Sciences, that provided data of 1179 patients from 2008 to 2018. The former and the later datasets consisted of
151 and 66 variables, respectively.

Data preparation
Following identifying common variables in the two datasets and a comprehensive review of relevant literature,
34 variables were selected, as outlined in Table 2. To augment patient data, the initial step involved gathering
specific values from the Health Information Systems at Tajrish Hospital in Tehran and Shahid Motahari Clinic
in Shiraz. In the second step, a total of 643 successful telephone calls were conducted to collect information on
patients’ survival status and lifestyle. Simultaneously, the survival status of patients who did not respond was
verified through the Iran Health Insurance System. Patients whose survival status could not be investigated
through any of these measures were subsequently excluded from the study. In total, data from 1875 patients
were utilized, comprising 741 individuals with less than a 5 year survival and 1134 individuals with a 5 year or
greater survival. Finally, the datasets were normalized and the missing values were managed using K-Nearest
Neighbors imputer.
The overall survival of patients was determined by calculating the time interval between the diagnosis and
the time of death. Specifically, if this interval exceeded 5 years or if the patient was alive with more than 5 years
having elapsed since the diagnosis of breast cancer, the label was assigned as 1; otherwise, it was labeled as 0.

Model development
DNN along with conventional machine learning models, such as LR, NB, KNN, DT, RF, Extra Trees, SVM,
Adaboost, GBoost, XGB, and MLP were used in this study. A brief explanation of each algorithm is provided
below:
DNN It is a type of artificial neural network with multiple hidden layers between the input and output
layers. DNNs are designed to automatically learn and model complex patterns by passing data through layered
architectures50.

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 2

www.nature.com/scientificreports/

Method
Feature Features Survival
Row References ML methods Tools Dataset selection (n) time Result
LR—LDA—LGBM—GBM— MSSQL
Nguyen et 6464 BC patients between 2008 and 2020 Literature AUC
1 RF—AdaBoost—XgBoost— server— 38 5 Years
al.42 in 3 center Review 0.95
ANN—voting ensemble Python
Othman et 1980 BC patients in an online dataset Accuracy
2 CNN—LSTM, GRU mRMR 25 5 Years
al.43 (METABRIC) 98%
Lotfnezhad
Accuracy
3 Afshar et C5.0—RIPPER Excel—R 856 BC patients between 2006 and 2012 15
84.42%
al.44
Lifelines— Cox Accuracy
4 Lou et al.45 DNN—KNN—SVM—NBC 1178 BC patients between 2007 and 2010 24 10 Years
Python regression 97.89%
Physicians
Ganggayah DT—RF—NN—SVM— Survey— Accuracy
5 R—Python 8066 BC patients between 1993 and 2016 23
et al.46 XGBoost—LR RF— 82.7%
VSURF
Forests of Accuracy
6 Kalafi et al.47 MLP—SVM—RF- DT Python 4092 BC patients between 1993 and 2017 23 5 Years
trees 88.2%
NB—RF—Adaboost—
Accuracy
7 Tapak et al.48 SVM—LSSVM—Adabag— R 550 BC patients between 1998 and 2013 RF 9
Ranged from 80 to 93%
LR
XGBoost—RF—SVM— Accuracy
8 Zhao et al.49 R 1874 BC patients between 2012 and 2016 KNN 27 5 Years
ANN—KNN Ranged from 69 to 73%
NB—TRF—1NN—
Montazeri et Weka Accuracy
9 Adaboost—SVM—RBFN— 900 BC patients between 1999 and 2007 1NN—TRF 8
al.40 software 96%
MLP

Table 1. Recent related studies. LR logistic regression, LDA linear discriminant analysis, LGBM light gradient
boosting machine, GBM gradient boosting machine, RF random forest, AdaBoost boosting, XGBoost extreme
gradient boosting, ANN artificial neural network, CNN convolutional neural network, LSTM long short-term
memory, GRU gated recurrent unit, RIPPER repeated incremental pruning to produce error reduction, DNN
deep neural network, KNN K-nearest neighbors, SVM support vector machine, NBC Naive Bayes classifier, DT
decision tree, NN neural network, VSURF variable selection using random forests, MLP multilayer perceptron,
NB Naive Bayes, LSSVM least squares support vector machine, TRF tree-based random forest, 1NN 1-nearest
neighbor, RBFN radial basis function network, mRMR minimum redundancy maximum relevance, AUC Area
under the curve

LR A statistical algorithm commonly applied to binary classification problems. It extends linear regression to
model the likelihood of a dichotomous outcome (e.g., occurrence vs. non-occurrence of an event) by mapping
predictions to probabilities51.
NB Bayes’ theorem is one of the fundamental principles in probability theory and mathematical statistics. In
this algorithm, variables are assumed to be independent of each other52.
KNN A non-parametric classification algorithm that determines the class of a test sample based on the classes
of its k nearest neighbors in the training data. The algorithm computes the distance between the test sample and
all training samples to find these neighbors53.
DT A predictive model that uses a tree-like structure to make decisions based on sequential tests of input
data. Each node represents a decision rule, and each branch represents the outcome of that rule, they ultimately
lead to prediction or classification54.
RF A machine learning algorithm that combines multiple decision trees to improve the prediction accuracy
and prevent overfitting. It operates by training each tree on a random subset of the data, with each tree providing
a “vote” for the outcome, and the most common vote across all trees is selected as the final prediction55.
Extra Trees An ensemble learning method used for classification and regression tasks, which improves
performance by aggregating predictions from multiple decision trees. Unlike traditional decision trees, Extra
Trees are built with more randomization in the tree creation process, notably by selecting random splits at each
node, which helps to avoid overfitting and enhances model accuracy56.
SVM A supervised machine learning algorithm works by finding the hyperplane that best separates different
data classes in a high-dimensional space. SVM is particularly effective in handling complex, non-linear data by
using kernel functions to transform data into higher dimensions, making it useful for accurate predictions57.
AdaBoost an ensemble machine learning technique that focuses on improving the performance of weak
classifiers by sequentially combining them into a stronger classifier58.
GBoost A machine learning algorithm that iteratively builds a model by training a sequence of weak learners,
typically decision trees, to correct the errors of previous ones59.
XGBoost A highly efficient machine learning algorithm based on gradient boosting principles, known for its
accuracy and speed in solving regression and classification problems. XGBoost also integrates several advanced
features like regularization, handling missing values, and parallelization, making it suitable for large datasets60.

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 3

www.nature.com/scientificreports/

Row Variables Mean Frequency

1 Age at diagnosis 49.44 years –
2 Marital status – (1) Single: 117, (2) Married: 1169, (3) Widow: 135, (4) Divorced: 41
3 Child – (1) Yes: 1215, (0) No: 192, Unknown: 489
4 Number of children 3.1 –
5 ER – (1) Positive: 1211, (2) Weak positive: 14, (3) Negative: 461, Unknown: 190
6 PR – (1) Positive: 1084, (2) Weak positive: 10, (3) Negative: 579, Unknown: 203
(1) Negative: 553
(2) 1 + : 353
7 HER2 – (3) 2 + : 254
(4) 3 + : 369
Unknown: 347
8 Tumor size 3.1 cm –
(1) 1: 211
(2) 2: 811
9 Tumor grade –
(3) 3: 454
Unknown: 400
10 Tumor type – (1) IDC: 1573, (2) DCIS: 71, (3) LIC: 67, (4) LCIS: 2, (5) Other: 71, Unknown: 92
11 LVI – (1) Yes: 829, (0) No: 774, Unknown: 273
12 PNI – (1) Yes: 571, (0) No: 666, Unknown: 639
13 Chemotherapy – (1) Yes: 1482, (0) No: 385, Unknown: 9
14 Chemotherapy type – (1) Neoadjuvant: 271, (2) Adjuvant: 940, (3) Both: 127, Unknown: 144
15 Chemotherapy sessions 8.14 –
16 Radiotherapy – (1) Yes: 1394, (0) No: 471, Unknown: 11
17 Radiotherapy type – (1) EBRT: 1210, (2) IORT: 74, (3) Both: 106, Unknown: 4
18 Radiotherapy Sessions 26.25 –
19 Lymph node Management – (1) ALND: 830, (2) SLNB: 362, (3) Both: 169, Unknown: 515
20 Surgery type – (1) BCS: 973, (2) MRM: 683, (3) Both: 95, (0) No Surgery: 39, Unknown: 86
21 Physical activity – (1) Yes: 654, (0) No: 1091, Unknown: 131
22 Menopause status – (1) Yes: 805, (0) No: 392, Unknown: 679
23 Menopause age 47 years –
24 Menstrual age 13.3 years –
25 Metastasis – (1) Yes: 707, (0) No: 943, Unknown: 226
26 Recurrence – (1) Yes: 158, (0) No: 1482, Unknown: 236
27 Smoking – (1) Yes: 21, (0) No: 1728, Unknown: 127
28 Alcohol Consumption – (1) Yes: 16, (0) No: 1732, Unknown: 128
29 Breastfeeding – (1) Yes: 1176, (0) No: 223, Unknown: 477
30 Number of Breastfed Children 2.96 –
31 Breastfeeding Month 57.24 months –
32 Laterality – (1) left: 943, (0) right: 830, (2) both: 28, Unknown: 75
34 Overall Survival – (1) at least 5 years: 1134, (0) less than 5 years: 741

Table 2. Study variables. Quantified labels are indicated in parentheses before each variable state.

MLP It is a type of artificial neural network composed of several layers of neurons, each of which processes
data through nonlinear activation functions. This architecture enables MLPs to learn complex patterns in the
data, making them suitable for classification and regression61.
The performance of these algorithms was thoroughly assessed and compared. Model development and
evaluation were conducted using the Python programming language, leveraging scikit-learn42,45,47, TensorFlow42,
and Autokeras libraries within the Jupyter Notebook environment.
Feature selection was conducted through three distinct approaches62,63. Initially, modeling was executed
using all features, which were chosen based on a review of relevant articles while also taking into account dataset
limitations. Subsequently, features were selected based on a two-tailed p value criterion (< 0.05) using scikit-
learn. Lastly, features were chosen through a survey methodology. To identify essential variables, a questionnaire
was formulated and completed by five oncology specialists, aiming to pinpoint essential features for the analysis.
As previously stated, the study involved modeling using a deep neural network and 11 conventional machine
learning models. To train and fit the conventional machine learning models, the dataset was initially divided into
two parts: the train set and the test set64. The train set served for model development, hyperparameter tuning,
and initial training. The train set was partitioned into five folds (k = 5), and the modeling process was iterated

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 4

www.nature.com/scientificreports/

five times. In each iteration, one of the folds served as the validation set, while the remaining four were used as
the train set. The model underwent training on the train set for each iteration and was subsequently validated on
the designated validation set. Different hyperparameters were employed during each fold, determined through
the Grid Search method. This method systematically explores a predefined range of hyperparameter values to
identify an optimal set that maximizes performance measures such as accuracy. Five distinct evaluation points,
or different accuracy values, were obtained for each iteration. The highest accuracy level signified the most
optimal model within that algorithm. As an example, Fig. 1 shows the DT algorithm training process, with
Shiraz datasets using k-fold and the accuracy score in each fold.
For training the deep neural network, the dataset was initially divided into two parts: the train set and the
test set. Subsequently, the Neural Architecture Search method, a technique in deep learning for automatically
exploring optimal neural network architectures, was applied. Hyperparameters were adjusted following this
exploration. To determine the optimal architecture and set hyperparameters, the Autokeras library was utilized.
Figure 2 illustrates the model development process employing both Deep Neural Network and conventional
machine learning techniques. Sets of hyperparameters and also the best parameters in each trained model are
attached as supplementary information.

Performance evaluation
At this stage, the trained models underwent evaluation on the test set to derive the ultimate estimate of
their performance on previously unseen data. The evaluation process encompassed two distinct approaches,
as illustrated in Fig. 1. Firstly, each model was assessed on a subset of the same dataset used for training,
constituting a cross-validation. Secondly, the evaluation was repeated on the dataset from the other center,
effectively constituting an external validation. These two evaluation methods provided a comprehensive
understanding of the models’ performance across both internal and external datasets65,66. It should be noted
that the hyperparameter setting was exclusively performed on the training dataset. In this study, the following
metrics were employed for evaluating the models.

1. Accuracy The percentage of people whose life status is correctly predicted.

T rue positive(T P )+T rue negative(T N )
 ccuracy =
A T otal
× 100
True positive (TP) indicates those individuals who are alive and are correctly predicted as alive. True negative
(TN) indicates those individuals who have died and are correctly considered dead.

2. Specificity The percentage of individuals who have died and are labeled 0 and are correctly considered dead.
T rue negative(T N )
Specif
 icity = F alse positive(F P )+T rue negative(T N )
× 100

Fig. 1. The DT algorithm training process and the accuracy score in each fold. Yellow boxes show the
validation set in each iteration.

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 5

www.nature.com/scientificreports/

Fig. 2. Dataset preparation and modeling process.

3. Sensitivity The percentage of individuals who are alive and have a label of 1 and are correctly predicted as
alive.
T rue positive(T P )
Sensitivity
 = T rue positive(T P )+F alse negative(F N )
× 100

4. Area under the Curve Indicates how well the model can distinguish between class labels and correctly pre-
dicts the model for classes 0 and 1.

Results
In this study, modeling was conducted using conventional machine learning algorithms and Deep Neural
Network in three distinct approaches. In the initial stage, modeling was executed using all available features
extracted from related articles. Subsequently, in the second and third stages, modeling occurred alongside
feature selection, taking into account p value and the opinions of 5 oncology experts, respectively. Given that
each algorithm was trained with three datasets—Tehran, Shiraz, and a combination of the two—it can be stated
that nine models were created for each algorithm. It is noteworthy that all models, except those trained using the
combined dataset, underwent evaluation through both cross-validation and external evaluation methods. Based
on the recorded evaluation metrics, the highest average accuracy in cross-validation was 94.29%, attributed to
models trained with the Shiraz dataset and features selected by oncologists. Additionally, the highest average
accuracy in external validation was 76.42%, observed in models trained with the Shiraz dataset and utilizing all
features. The maximum average AUC in cross-validation, reaching 0.983, was associated with models trained
using the Shiraz dataset and features selected based on the p-value. In external validation, this AUC value was
0.851, achieved by models trained with the Tehran dataset and features selected by oncologists. The subsequent
section provides a detailed presentation of the evaluation results.

Evaluation and performance comparison of trained models with all features

According to Table 3, the highest accuracy achieved on the test data in the cross-validation was 95.43%, which
belongs to the GBoost and Extra Tree models, which were trained with the Shiraz dataset. Additionally, the
highest accuracy level in external validation reached 85.56%, which was achieved using the DNN model. Figure
3 illustrates the architecture of this DNN model. Among the conventional models, the highest accuracy was
81.69%, obtained with the SVM model. Both models were trained with the Shiraz dataset and tested on the

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 6

www.nature.com/scientificreports/

Models Train dataset Train accuracy Test dataset Test accuracy AUC Sensitivity Specificity
Shiraz (cross) 93.77 0.986 93.8 93.75
Shiraz 94.49
Tehran (external) 79.91 0.789 81.71 65.78
KNN Tehran (cross) 87.4 0.767 100 10.52
Tehran 97.76
Shiraz (external) 45.22 0.921 100 0
Combined 92.93 Combined (cross) 86.66 0.889 88.16 83.84
Shiraz (cross) 93.36 0.966 90.26 96.09
Shiraz 92.41
Tehran (external) 73.21 0.641 77.85 36.84
NB Tehran (cross) 75.55 0.708 84.48 21.05
Tehran 85.1
Shiraz (external) 49.12 0.770 94.42 12.48
Combined 89.33 Combined (cross) 86.93 0.912 89.38 82.3
Shiraz (cross) 94.6 0.966 92.03 96.87
Shiraz 95.94
Tehran (external) 68.6 0.717 68.12 72.36
DT Tehran (cross) 86.67 0.673 97.41 21.05
Tehran 91.43
Shiraz (external) 46.05 0.499 99.25 3
Combined 93.26 Combined (cross) 86.67 0.890 86.94 86.15
Shiraz (cross) 93.36 0.984 92.03 94.53
Shiraz 94.9
Tehran (external) 78.57 0.792 80.37 64.47
LR Tehran (cross) 85.18 0.765 98.27 5.26
Tehran 90.68
Shiraz (external) 46.79 0.954 99.25 4.36
Combined 91.86 Combined (cross) 87.2 0.938 88.16 85.38
Shiraz (cross) 95.02 0.991 92.03 97.65
Shiraz 96.05
Tehran (external) 69.04 0.775 68.62 72.36
RF Tehran (cross) 86.66 0.784 100 5.26
Tehran 94..22
Shiraz (external) 44.72 0.935 100 0
Combined 98.73 Combined (cross) 89.6 0.958 91.83 85.38
Shiraz (cross) 95.43 0.994 92.92 97.65
Shiraz 99.89
Tehran (external) 68.6 0.790 67.44 77.63
Extra Trees Tehran (cross) 87.4 0.810 100 10.52
Tehran 96.46
Shiraz (external) 44.72 0.960 100 0
Combined 99.86 Combined (cross) 87.73 0.956 88.97 85.38
Shiraz (cross) 94.6 0.991 92.03 96.87
Shiraz 96.36
Tehran (external) 81.69 0.805 83.89 64.47
SVM Tehran (cross) 86.66 0.772 100 5.26
Tehran 94.59
Shiraz (external) 44.72 0.934 100 0
Combined 96.26 Combined (cross) 87.73 0.937 88.97 85.38
Shiraz (cross) 95.02 0.984 92.03 97.65
Shiraz 95.63
Tehran (external) 69.34 0.753 68.95 72.36
AdaBoost Tehran (cross) 85.92 0.730 98.27 10.52
Tehran 99.44
Shiraz (external) 48.29 0.538 99.25 07.06
Combined 92.06 Combined (cross) 86.93 0.948 87.34 86.15
Shiraz (cross) 95.43 0.995 92.92 97.65
Shiraz 95.73
Tehran (external) 69.34 0.751 68.95 72.36
GBoost Tehran (cross) 86.66 0.769 100 5.26
Tehran 93.48
Shiraz (external) 44.97 0.937 99.81 0.6
Combined 93.66 Combined (cross) 87.46 0.949 87.75 86.92
Shiraz (cross) 95.85 0.994 95.57 96.09
Shiraz 99.89
Tehran (external) 67.26 0.749 66.27 75
XGB Tehran (cross) 90.37 0.854 100 31.57
Tehran 98.32
Shiraz (external) 46.63 0.908 99.25 04.06
Combined 96.06 Combined (cross) 88.8 0.957 90.2 86.15
Shiraz (cross) 93.36 0.983 92.92 93.75
Shiraz 95.01
Tehran (external) 78.86 0.806 80.7 64.47
MLP Tehran (cross) 88.14 0.762 99.13 10.52
Tehran 91.06
Shiraz (external) 46.05 0.956 99.44 2.85
Combined 91.93 Combined (cross) 87.46 0.944 88.57 85.38
Continued

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 7

www.nature.com/scientificreports/

Models Train dataset Train accuracy Test dataset Test accuracy AUC Sensitivity Specificity
Shiraz (cross) 89.62 0.958 88.49 90.62
Shiraz 94.38
Tehran (external) 85.56 0.712 43.42 90.93
DNN Tehran (cross) 85.92 0.738 96.55 21.05
Tehran 93.11
Shiraz (external) 44.72 0.540 99.62 0
Combined 95.2 Combined (cross) 87.2 0.930 89.38 83.07

Table 3. Evaluation results of trained models based on all features.

Fig. 3. The Architecture of the DNN model that trained with the Shiraz dataset and showed the highest
accuracy in external validation among all 108 trained models.

Tehran dataset. Figure 4 shows the learning curves of the models that recorded the highest cross or external
validation accuracy. The pinnacle AUC levels in cross-validation were attained by XGB and Extra Trees with
0.994 and in external validation by Extra Trees with 0.960. Figure 5 illustrates the ROC curves of models that
recorded the highest AUC in cross and external validation.

Evaluation and performance comparison of trained models with selected features based on P
value
In this part, feature selection was done based on two-tailed p value (< 0.05) before modeling. The selected
features based on Shiraz, Tehran, and combined datasets are shown in Table 4.
According to Table 5, the highest accuracy achieved on the test data in cross-validation was 96.26%, which
belongs to the DT model trained with the Shiraz dataset. Moreover, the highest level of accuracy in external
validation was 82.89%, and it was obtained using the SVM model, which was trained with the Shiraz dataset and
tested on the Tehran dataset. Figure 6 shows the learning curves of the models that recorded the highest cross or
external validation accuracy. The highest level of AUC in cross and external validation was obtained using Extra
Trees with 0.992 and MLP with 0.944, respectively. Figure 7 illustrates the ROC curves of models that recorded
the highest AUC in cross and external validation.

Evaluation and performance comparison of trained models with features selected by

oncologists
In this part, preceding the modeling process, feature selection was carried out based on a questionnaire
completed by five oncology specialists. The questionnaire assigned importance and impact scores to each feature
on the survival of breast cancer patients, ranging from 1 to 5 (1: unimportant, 5: very important). Features with
an average score greater than or equal to 3 were selected for use in modeling. Figure 8 illustrates the selected

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 8

www.nature.com/scientificreports/

Fig. 4. Learning curves for Extra Trees, SVM, GBoost, and DNN models trained on the Shiraz datasets with all
features. The Extra Trees learning curve indicates that the training score does not improve with more training
data, but the cross-validation score does. The SVM and GBoost learning curves show that while the training
scores decrease with more training data, the cross-validation scores increase. Using all the training data makes
the training and cross-validation scores more reasonable and realistic. The DNN learning curve indicates that
both the training and cross-validation scores increase with more training data.

features along with their respective average scores. It is noteworthy that the features selected in Shiraz and
Tehran and the combined datasets were the same.
As per Table 6, the highest accuracy on the test data in cross-validation reached 95.85%, which is attributed to
the GBoost model trained with the Shiraz dataset. Additionally, the highest accuracy in external validation was
81.54%, achieved by the DNN model. The highest accuracy of the conventional machine learning models was
77.82%, obtained with the LR model. Both the DNN and LR models were trained using the Shiraz dataset and
tested on the Tehran dataset. Figure 9 shows the learning curves of the models that recorded the highest cross or
external validation accuracy. The maximum AUC levels in cross-validation and external validation were attained
by RF with 0.992 and LR with 0.972, respectively. Figure 10 illustrates the ROC curves of models that recorded
the highest AUC in cross and external validation.

Discussion
In this study, the utilization of two datasets significantly mitigates the likelihood of bias associated with single-
center studies, an issue often cited in similar research 45. Moreover, each algorithm underwent training three
times: once with all features, once with features selected based on p value, and once again with features selected
by oncologists. This resulted in the development of a total of 108 models. Subsequently, the performance of
conventional machine learning models and DNN was compared.

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 9

www.nature.com/scientificreports/

Fig. 5. ROC curves for Extra Trees performance in cross and external validation and XGB in cross-validation.

Dataset Selected features and P values

Metastasis (0), Recurrence (0.021), Age at diagnosis (0.016), ER (0), PR (0), Chemotherapy (0), Chemotherapy type (0), Chemotherapy sessions
Shiraz (0), Tumor size (0), LVI (0), PNI (0), Lymph node management (0), Radiotherapy (0), Radiotherapy type (0), Radiotherapy sessions (0), Surgery
Type (0), physical activity (0), Menopause status (0), Menopause age (0.013), Number of Children (0.003)
Metastasis (0), Recurrence (0), ER (0), PR (0.018), Tumor size (0.031), LVI (0.001), Surgery Type (0.004), Smoking (0.013), Alcohol
Tehran
Consumption (0.014), Menopause status (0.030), Breastfeeding Month (0.029)
Metastasis (0), Age at diagnosis (0.006), ER (0), PR (0), HER2 (0.006), Chemotherapy (0), Chemotherapy type (0), Chemotherapy sessions (0),
Combined Tumor size (0), LVI (0), PNI (0), Lymph node management (0), Radiotherapy (0), Radiotherapy type (0), Radiotherapy sessions (0), Surgery
Type (0), Physical Activity (0), Menopause age (0.024), Number of Children (0), Number of Breastfed Children (0), Breastfeeding Month (0.006)

Table 4. Selected features in each dataset based on P value.

Certainly, one of the strengths of this study lies in the comprehensive consideration of various variables
encompassing tumor characteristics, tumor markers, patient clinical information, patient characteristics, and
lifestyle factors. As evident, an essential step before modeling is feature selection, a task accomplished through
various methods such as RF, 1NN, KNN, and Cox regression. Beyond these techniques, another approach to
feature selection involves consulting experts. Study 42 emphasizes the importance of specialist input in refining
the feature set. Initially, their dataset contained 113 features, but after expert consultation, 89 features were
discarded. As previously mentioned, in our study, feature selection was conducted not only based on p value but
also through a survey involving oncologists. As a result of this expert input, 24 items were selected from the initial
set of 32 features. In this study’s overall findings of feature selection methods, common features that emerged
as significant include metastasis, recurrence, age at diagnosis, estrogen and progesterone hormone receptors,
tumor size, lymph vascular invasion, and the type of surgery performed. It is noteworthy that tumor size, age
at diagnosis, hormone receptors, and surgery have consistently been identified as important characteristics in
many studies, aligning with the results observed in this investigation46,47,49.
The results reveal that the Extra Trees and GBoost models, trained on all features, achieved the highest
cross-validation accuracy at 95.43%. Moreover, the DNN model demonstrated the highest external validation
accuracy at 85.56%. This finding is consistent with the results of study45 that both cross and external validation
were conducted. According to this study45 the deep neural network model outperformed in all evaluation
indicators. In contrast, among our models trained with all features, the XGB model demonstrated the highest
AUC in cross-validation, whereas in study 45 there was not a significant performance difference between XGB
and other models.
For models trained on features selected based on p value, the DT model achieved the highest cross-validation
accuracy at 96.26%. However, this contrasts with study43, which reported negligible performance differences
between DT, SVM, and RF models. In external validation, the SVM model’s accuracy of 82.89% supports
findings from study41, which also recognized SVM’s strong performance after DNN. The highest amount of
AUC among trained models with selected features based on P value in cross and external validations was related
to Extra Trees and MLP, respectively. Meanwhile, based on the results of study40, the AUC of the MLP model was
not much different from other models.
Regarding models trained on features selected by oncologists, the highest accuracy of cross-validation was
95.85% and was related to the GBoost model. Besides, the highest accuracy of external validation was 81.59%
and related to the DNN. This result is consistent with study41, which highlighted DNN’s superior performance in
external validation. However, study46, where feature selection was conducted with the assistance of expertise, the
RF model demonstrated the highest level of performance accuracy. The highest AUCs for models trained with

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 10

www.nature.com/scientificreports/

Models Train dataset Train accuracy Test dataset Test accuracy AUC Sensitivity Specificity
Shiraz (cross) 93.36 0.985 92.92 93.75
Shiraz 94.38
Tehran (external) 82.74 0.763 85.91 57.89
KNN Tehran (cross) 87.4 0.759 100 10.52
Tehran 100
Shiraz (external) 44.72 0.852 100 0
Combined 92.13 Combined (cross) 86.66 0.936 87.34 85.38
Shiraz (cross) 93.77 0.969 89.38 97.65
Shiraz 92.31
Tehran (external) 68.45 0.712 68.46 68.42
NB Tehran (cross) 83.7 0.571 97.41 0
Tehran 85.28
Shiraz (external) 47.29 0.750 95.91 7.96
Combined 88.6 Combined (cross) 85.33 0.903 90.2 76.15
Shiraz (cross) 96.26 0.971 93.8 98.43
Shiraz 95.84
Tehran (external) 65.62 0.738 63.76 80.26
DT Tehran (cross) 91.11 0.836 100 36.84
Tehran 90.5
Shiraz (external) 64.75 0.896 98.32 37.6
Combined 91.87 Combined (cross) 85.6 0.920 85.3 86.15
Shiraz (cross) 93.77 0.984 92.03 95.31
Shiraz 95.01
Tehran (external) 81.99 0.781 85.07 57.89
LR Tehran (cross) 88.15 0.780 100 15.78
Tehran 91.24
Shiraz (external) 47.05 0.930 99.25 4.81
Combined 91.13 Combined (cross) 88 0.935 91.42 81.53
Shiraz (cross) 93.77 0.988 92.03 95.31
Shiraz 96.78
Tehran (external) 69.64 0.767 69.46 71.05
RF Tehran (cross) 88.14 0.798 100 15.78
Tehran 91.99
Shiraz (external) 45.13 0.936 99.81 0.9
Combined 99.26 Combined (cross) 90.13 0.944 93.46 83.84
Shiraz (cross) 94.6 0.992 92.03 96.87
Shiraz 96.77
Tehran (external) 77.08 0.775 78.69 64.47
Extra Trees Tehran (cross) 87.4 0.772 100 10.52
Tehran 100
Shiraz (external) 44.72 0.928 99.81 0
Combined 100 Combined (cross) 88.53 0.948 91.02 83.84
Shiraz (cross) 93.36 0.987 92.03 94.53
Shiraz 96.57
Tehran (external) 82.89 0.771 85.4 63.16
SVM Tehran (cross) 88.14 0.798 99.13 21.05
Tehran 91.62
Shiraz (external) 46.96 0.884 99.62 4.36
Combined 93.06 Combined (cross) 86.4 0.942 86.93 85.38
Shiraz (cross) 94.6 0.981 92.03 96.87
Shiraz 95.74
Tehran (external) 69.64 0.736 69.46 71.05
AdaBoost Tehran (cross) 85.92 0.584 100 0
Tehran 93.29
Shiraz (external) 45.71 0.427 99.81 1.95
Combined 92.2 Combined (cross) 87.73 0.949 87.34 88.46
Shiraz (cross) 94.6 0.987 92.03 96.87
Shiraz 95.74
Tehran (external) 69.64 0.736 69.46 71.05
GBoost Tehran (cross) 88.14 0.819 99.13 21.05
Tehran 93.29
Shiraz (external) 47.21 0.882 99.44 4.96
Combined 100 Combined (cross) 86.13 0.914 88.97 80.76
Shiraz (cross) 94.6 0.989 92.92 96.09
Shiraz 98.54
Tehran (external) 67.71 0.732 66.44 77.63
XGB Tehran (cross) 88.88 0.803 99.13 26.31
Tehran 94.04
Shiraz (external) 45.96 0.924 99.44 02.07
Combined 99.53 Combined (cross) 88.26 0.946 89.79 85.38
Shiraz (cross) 93.77 0.984 92.03 95.31
Shiraz 94.91
Tehran (external) 81.99 0.783 84.73 60.53
MLP Tehran (cross) 87.4 0.780 98.27 21.05
Tehran 90.68
Shiraz (external) 50.37 0.944 98.69 11.27
Combined 97 Combined (cross) 87.2 0.928 82.24 77.69
Continued

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 11

www.nature.com/scientificreports/

Models Train dataset Train accuracy Test dataset Test accuracy AUC Sensitivity Specificity
Shiraz (cross) 83.81 0.975 92.94 94.59
Shiraz 93.86
Tehran (external) 75.59 0.646 79.02 48.68
DNN Tehran (cross) 85.18 0.905 99.37 22.08
Tehran 91.24
Shiraz (external) 44.97 0.667 0.4 100
Combined 91.73 Combined (cross) 82.93 0.894 95.91 58.46

Table 5. Evaluation results of trained models based on selected features (P value).

Fig. 6. Learning curves for DT and SVM models trained on the Shiraz datasets with features selected based
on p values. The DT learning curve indicates that, despite fluctuations, both the training and cross-validation
scores increase with more training data. The SVM learning curve shows that while the training scores decrease
with more training data, the cross-validation scores increase, and using all the training data makes the training
and cross-validation scores more reasonable and realistic.

Fig. 7. ROC curves for MLP performance in external validation and Extra Trees in cross-validation.

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 12

www.nature.com/scientificreports/

Fig. 8. Selected features and their average scores determined by oncologists.

oncologist-selected features were achieved by RF in cross-validation and LR in external validation, contrasting

with the findings of study42, where XGB recorded the highest AUC in cross-validation.
The findings of the current study are in line with study41, which reported the DNN as the best model. Overall,
aside from DNN, other studies36,38,43,45 indicated that models based on neural networks such as ANN and MLP,
showed better performance. It is noteworthy to say that the mentioned contractions in studies36,42,43,45 could be
assigned to utilizing different datasets as well as feature selection approaches.

Study limitations and future considerations

This study encountered several limitations. Firstly, the inaccessibility of patients’ genetic data was a significant
constraint. Combining genetic data with other available information could potentially enhance the efficiency of
the models. Additionally, the absence of certain aspects of the patients’ medical history, such as blood pressure,
blood sugar levels, and other cancers, could notably impact the performance and accuracy of both ML and DL
models. Certainly, considering the drugs used in the treatment process would be a valuable addition, bringing
the study results closer to the outcomes obtained in real-world scenarios. Future investigations could include
using medical images in addition to other forms of data and training deep learning models like CNN. Moreover,
developing and comparing metastasis and recurrence prediction models could provide a broader perspective
in this field. Furthermore, incorporating online and accessible datasets for external validation could also be
practical to enhance the applicability of our findings.

Conclusion
To the best of our knowledge, this study represents a pioneering effort in Iran, being the first to introduce a survival
prediction model using deep learning. Leveraging data from two centers and incorporating external validation
further distinguishes the study. The results indicate that, overall, the DNN model demonstrated superior
prediction accuracy in external validation. This could be because DNN can capture non-linear relationships and
interactions among features better than simpler models. However, DNN was not consistently at a higher level
in other performance metrics. Moreover, among the conventional models, SVM showed the highest prediction
accuracy in external validation. The reason behind this could be that SVM employs kernel functions to transform
data into higher dimensions, allowing it to capture complex relationships between features. Notably, evaluation
metrics were generally higher for models trained with the Shiraz dataset. This discrepancy might be attributed
to the fewer missing values in the Shiraz dataset compared to the Tehran dataset, which was addressed using the
KNN algorithm. With an increasing number of similar studies and positive outcomes, there is optimism that
ongoing advancements in the field will lead to optimized medical decisions and improved disease prognosis
through the utilization of deep learning algorithms to uncover hidden patterns in data.

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 13

www.nature.com/scientificreports/

Models Train dataset Train accuracy Test dataset Test accuracy AUC Sensitivity Specificity
Shiraz (cross) 93.77 0.985 92.03 95.31
Shiraz 94.28
Tehran (external) 75 0.780 76.17 65.78
KNN Tehran (cross) 88.14 0.660 100 15.78
Tehran 91.24
Shiraz (external) 45.71 0.904 99.81 1.95
Combined 92.6 Combined (cross) 88.8 0.927 89.38 87.69
Shiraz (cross) 93.77 0.960 90.26 96.87
Shiraz 92.2
Tehran (external) 73.21 0.658 77.18 42.1
NB Tehran (cross) 77.03 0.697 83.62 36.84
Tehran 84.72
Shiraz (external) 54.03 0.819 98.69 17.89
Combined 89.46 Combined (cross) 88.26 0.916 89.38 86.15
Shiraz (cross) 95.43 0.976 92.92 97.65
Shiraz 95.11
Tehran (external) 66.51 0.728 65.43 75
DT Tehran (cross) 84.44 0.636 96.55 10.52
Tehran 89.75
Shiraz (external) 52.95 0.815 99.07 15.63
Combined 87.4 Combined (cross) 81.6 0.895 76.32 91.53
Shiraz (cross) 93.77 0.984 92.03 95.31
Shiraz 95.01
Tehran (external) 77.82 0.790 79.36 65.78
LR Tehran (cross) 88.14 0.775 100 15.78
Tehran 90.13
Shiraz (external) 54.94 0.972 99.81 18.64
Combined 91.06 Combined (cross) 87.46 0.943 88.16 86.15
Shiraz (cross) 95.02 0.992 92.03 97.65
Shiraz 99.37
Tehran (external) 68.6 0.771 68.12 72.36
RF Tehran (cross) 88.14 0.744 100 15.78
Tehran 97.02
Shiraz (external) 49.62 0.829 98.88 9.77
Combined 96.8 Combined (cross) 89.33 0.948 90.2 87.69
Shiraz (cross) 95.02 0.991 92.03 97.65
Shiraz 95.94
Tehran (external) 69.04 0.800 68.62 72.36
Extra Trees Tehran (cross) 88.14 0.765 100 15.78
Tehran 99.44
Shiraz (external) 44.88 0.959 99.81 0.4
Combined 99.4 Combined (cross) 88.26 0.949 88.97 86.92
Shiraz (cross) 93.77 0.986 92.03 95.31
Shiraz 95.21
Tehran (external) 77.23 0.807 78.02 71.05
SVM Tehran (cross) 88.14 0.732 100 15.78
Tehran 91.99
Shiraz (external) 50.7 0.837 99.44 11.27
Combined 95.13 Combined (cross) 87.46 0.942 88.57 85.38
Shiraz (cross) 95.02 0.984 92.03 97.65
Shiraz 95.53
Tehran (external) 69.04 0.755 68.62 72.36
AdaBoost Tehran (cross) 86.66 0.743 100 5.26
Tehran 89.75
Shiraz (external) 47.46 0.873 99.25 5.56
Combined 95.33 Combined (cross) 88 0.931 89.79 84.61
Shiraz (cross) 95.85 0.989 94.69 94.87
Shiraz 98.12
Tehran (external) 68.45 0.766 67.78 73.68
GBoost Tehran (cross) 87.4 0.780 100 10.52
Tehran 91.06
Shiraz (external) 44.97 0.939 99.81 0
Combined 99.53 Combined (cross) 88.26 0.945 88.97 86.92
Shiraz (cross) 95.02 0.990 93.8 96.09
Shiraz 98.23
Tehran (external) 65.77 0.765 63.92 80.26
XGB Tehran (cross) 88.88 0.790 100 21.05
Tehran 97.95
Shiraz (external) 46.3 0.938 99.62 3.15
Combined 94.33 Combined (cross) 87.73 0.949 88.97 85.38
Shiraz (cross) 93.77 0.983 92.03 95.31
Shiraz 95.11
Tehran (external) 74.25 0.804 74.32 73.68
MLP Tehran (cross) 85.18 0.768 98.27 10.52
Tehran 90.68
Shiraz (external) 52.45 0.957 99.81 14.13
Combined 91.6 Combined (cross) 87.46 0.943 87.34 86.92
Continued

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 14

www.nature.com/scientificreports/

Models Train dataset Train accuracy Test dataset Test accuracy AUC Sensitivity Specificity
Shiraz (cross) 91.28 0.967 92.94 95.9
Shiraz 94.59
Tehran (external) 81.54 0.761 85.57 50
DNN Tehran (cross) 85.92 0.873 100 8.77
Tehran 90.31
Shiraz (external) 52.72 0.376 100 0
Combined 93.33 Combined (cross) 87.73 0.933 89.79 83.84

Table 6. Evaluation results of trained models based on features selected by oncologists.

Fig. 9. Learning curves for LR, GBoost, and DNN models trained on the Shiraz datasets with features selected
by oncologists. The LR and DNN learning curves indicate that despite fluctuations, both the training and cross-
validation scores increase with more training data. The GBoost learning curve shows that while the training
score decreases with more training data, the cross-validation score increases, and using all the training data
makes the training and cross-validation scores more reasonable and realistic.

Fig. 10. ROC curves for LR performance in external validation and RF in cross-validation.

Data availability
The data that support the findings of this study are not openly available due to the policies and regulations of the
data centers that provided data. The raw data could be made available from the corresponding authors upon rea-
sonable request. The Python source code is available at https://doi.​org/ht​tps://doi.or​g/10.5281/z​enodo.12805879.

Received: 28 January 2024; Accepted: 28 November 2024

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 15

www.nature.com/scientificreports/

References
1. Fitzmaurice, C. et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and
disability-adjusted life-years for 32 cancer groups, 1990 to 2015: A systematic analysis for the global burden of disease study. JAMA
Oncol. 3(4), 524–548 (2017).
2. Łukasiewicz, S. et al. Breast cancer—Epidemiology, risk factors, classification, prognostic markers, and current treatment
strategies—An updated review. Cancers 13(17), 4287 (2021).
3. Lei, S. et al. Global patterns of breast cancer incidence and mortality: A population-based cancer registry data analysis from 2000
to 2020. Cancer Commun. 41(11), 1183–1194 (2021).
4. Gorgzadeh, A. et al. Investigating the properties and cytotoxicity of cisplatin-loaded nano-polybutylcyanoacrylate on breast cancer
cells. Asian Pac. J. Cancer Biol. 8(4), 345–350 (2023).
5. WHO. Breast cancer description available from: https://www.​who.int/news​-room/fact-s​heets/detai​l/breast-cancer (2020).
6. Ahmad, A., Breast cancer statistics: recent trends. Breast cancer metastasis and drug resistance: challenges and progress, pp. 1–7
(2019).
7. CDC. Breast Cancer Statistics avaavailable from: https://www.​cdc.gov/canc​er/breast/st​atistics/in​dex.htm (2023).
8. Taylor, C. et al. Breast cancer mortality in 500 000 women with early invasive breast cancer in England, 1993–2015: Population
based observational cohort study. Bmj https://doi.org/10.1136/bmj-2022-074684 (2023).
9. Aryannejad, A. et al. National and subnational burden of female and male breast cancer and risk factors in Iran from 1990 to 2019:
Results from the global burden of disease study 2019. Breast Cancer Res. 25(1), 47 (2023).
10. Rahimzadeh, S. et al. Geographical and socioeconomic inequalities in female breast cancer incidence and mortality in Iran: A
Bayesian spatial analysis of registry data. PLoS ONE 16(3), e0248723 (2021).
11. Alizadeh, M. et al. Age at diagnosis of breast cancer in Iran: A systematic review and meta-analysis. Iran. J. Public Health 50(8),
1564 (2021).
12. Akbari, M. E. et al. Ten-year survival of breast cancer in Iran: A national study (retrospective cohort study). Breast Care (Basel)
18(1), 12–21 (2023).
13. Ginsburg, O. et al. Breast cancer early detection: A phased approach to implementation. Cancer 126(S10), 2379–2393 (2020).
14. Maajani, K. et al. The global and regional survival rate of women with breast cancer: A systematic review and meta-analysis. Clin.
Breast Cancer 19(3), 165–177 (2019).
15. Denfeld, Q. E., Burger, D. & Lee, C. S. Survival analysis 101: An easy start guide to analysing time-to-event data. Eur. J. Cardiovasc.
Nurs. 22(3), 332–337 (2023).
16. Ghaderzadeh, M. & Aria, M. Management of Covid-19 detection using artificial intelligence in 2020 pandemic. In Proceedings
of the 5th International Conference on Medical and Health Informatics. Association for Computing Machinery: Kyoto, Japan, pp.
32–38 (2021).
17. Aria, M., Ghaderzadeh, M. & Asadi, F. X-ray equipped with artificial intelligence: Changing the COVID-19 diagnostic paradigm
during the pandemic. BioMed. Res. Int. 2021, 9942873 (2021).
18. Rai, S., Mishra, P. & Ghoshal, U. C. Survival analysis: A primer for the clinician scientists. Indian J. Gastroenterol. 40(5), 541–549
(2021).
19. Indrayan, A. & Tripathi, C. B. Survival analysis: Where, why, what and how?. Indian Pediatr. 59(1), 74–79 (2022).
20. Wongvibulsin, S., Wu, K. C. & Zeger, S. L. Clinical risk prediction with random forests for survival, longitudinal, and multivariate
(RF-SLAM) data analysis. BMC Med. Res. Methodol. 20(1), 1 (2019).
21. Fraisse, J. et al. Optimal biological dose: A systematic review in cancer phase I clinical trials. BMC Cancer 21, 1–10 (2021).
22. Lotfnezhad Afshar, H. et al. Prediction of breast cancer survival through knowledge discovery in databases. Glob. J Health Sci. 7(4),
392–398 (2015).
23. Akgün, C. et al. Prognostic factors affecting survival in breast cancer patients age 40 or younger. J. Exp. Clin. Med. 39(4), 928–933
(2022).
24. Escala-Garcia, M. et al. Breast cancer risk factors and their effects on survival: A Mendelian randomisation study. BMC Med. 18(1),
327 (2020).
25. Arefinia, F. et al. Non-invasive fractional flow reserve estimation using deep learning on intermediate left anterior descending
coronary artery lesion angiography images. Sci. Rep. 14(1), 1818 (2024).
26. Pacal, İ. Deep learning approaches for classification of breast cancer in ultrasound (US) images. J. Inst. Sci. Technol. 12(4), 1917–
1927 (2022).
27. Işık, G. & Paçal, İ. Few-shot classification of ultrasound breast cancer images using meta-learning algorithms. Neural Comput.
Appl. 36(20), 12047–12059 (2024).
28. Coşkun, D. et al. A comparative study of YOLO models and a transformer-based YOLOv5 model for mass detection in
mammograms. Turk. J. Electr. Eng. Comput. Sci. 31(7), 1294–1313 (2023).
29. Shimizu, H. & Nakayama, K. I. Artificial intelligence in oncology. Cancer Sci. 111(5), 1452–1460 (2020).
30. Zarean Shahraki, S. et al. Time-related survival prediction in molecular subtypes of breast cancer using time-to-event deep-
learning-based models. Front. Oncol. 13, 1147604 (2023).
31. Tomatis, S. et al. Late rectal bleeding after 3D-CRT for prostate cancer: Development of a neural-network-based predictive model.
Phys. Med. Biol. 57(5), 1399 (2012).
32. Tran, K. A. et al. Deep learning in cancer diagnosis, prognosis and treatment selection. Genome Med. 13(1), 152 (2021).
33. Ghaderzadeh, M. et al. A fast and efficient CNN model for B-ALL diagnosis and its subtypes classification using peripheral blood
smear images. Int. J. Intell. Syst. 37(8), 5113–5133 (2022).
34. Bayani, A. et al. Identifying predictors of varices grading in patients with cirrhosis using ensemble learning. Clin. Chem. Lab. Med.
(CCLM) 60(12), 1938–1945 (2022).
35. Bayani, A. et al. Performance of machine learning techniques on prediction of esophageal varices grades among patients with
cirrhosis. Clin. Chem. Lab. Med. (CCLM) 60(12), 1955–1962 (2022).
36. Ghaderzadeh, M. et al. Deep convolutional neural network-based computer-aided detection system for COVID-19 using multiple
lung scans: design and implementation study. J. Med. Internet Res. 23(4), e27468 (2021).
37. Boldrini, L. et al. Deep learning: a review for the radiation oncologist. Front. Oncol. 9, 977 (2019).
38. Mihaylov, I., Nisheva, M. & Vassilev, D. Application of machine learning models for survival prognosis in breast cancer studies.
Information 10(3), 93 (2019).
39. Arya, N. & Saha, S. Multi-modal advanced deep learning architectures for breast cancer survival prediction. Knowl. -Based Syst.
221, 106965 (2021).
40. Montazeri, M. et al. Machine learning models in breast cancer survival prediction. Technol. Health Care 24, 31–42 (2016).
41. Yang, P.-T. et al. Breast cancer recurrence prediction with ensemble methods and cost-sensitive learning. Open Med. 16(1), 754–
768 (2021).
42. Nguyen, Q. T. N. et al. Machine learning approaches for predicting 5 year breast cancer survival: A multicenter study. Cancer Sci.
114(10), 4063–4072 (2023).
43. Othman, N. A., Abdel-Fattah, M. A. & Ali, A. T. A hybrid deep learning framework with decision-level fusion for breast cancer
survival prediction. Big Data Cognit. Comput. 7(1), 50 (2023).
44. Lotfnezhad Afshar, H. et al. Prediction of breast cancer survival by machine learning methods: An application of multiple
imputation. Iran J. Public Health 50(3), 598–605 (2021).

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 16

www.nature.com/scientificreports/

45. Lou, S. J. et al. Breast cancer surgery 10 year survival prediction by machine learning: A large prospective cohort study. Biology
(Basel) 11(1), 47 (2021).
46. Ganggayah, M. D. et al. Predicting factors for survival of breast cancer patients using machine learning techniques. BMC Med.
Inform. Decis. Mak. 19(1), 48 (2019).
47. Kalafi, E. et al. Machine learning and deep learning approaches in breast cancer survival prediction using clinical data. Folia Boil.
65(5/6), 212–220 (2019).
48. Tapak, L. et al. Prediction of survival and metastasis in breast cancer patients using machine learning classifiers. Clin. Epidemiol.
Glob. Health 7(3), 293–299 (2019).
49. Zhao, M. et al. Machine learning with k-means dimensional reduction for predicting survival outcomes in patients with breast
cancer. Cancer Inform. 17, 1176935118810215 (2018).
50. Shrestha, A. & Mahmood, A. Review of deep learning algorithms and architectures. IEEE Access 7, 53040–53065 (2019).
51. Domínguez-Rodríguez, S. et al. Machine learning outperformed logistic regression classification even with limit sample size: A
model to predict pediatric HIV mortality and clinical progression to AIDS. PLOS ONE 17(10), e0276116 (2022).
52. Chen, H. et al. Improved naive Bayes classification algorithm for traffic risk management. EURASIP J. Adv. Signal Process. 2021(1),
30 (2021).
53. Saadatfar, H. et al. A new k-nearest neighbors classifier for big data based on efficient data pruning. Mathematics 8, 286. ​h​t​t​p​s​:​/​/d
​ ​
o​i​.​o​r​g​/​1​0.​ ​3​3​9​0​/​m​a​t​h8​ ​0​2​0​2​8​6​​​​ (2020).
54. Blockeel, H. et al. Decision trees: from efficient prediction to responsible AI. Front. Artif. Intell. 6, 1124553 (2023).
55. Hu, L. & Li, L. Using tree-based machine learning for health studies: Literature review and case series. Int. J. Environ. Res. Public
Health 19(23), 16080. https://doi.org/10.3390/ijerph192316080 (2022).
56. Akinola, S., Leelakrishna, R. & Varadarajan, V. Enhancing cardiovascular disease prediction: A hybrid machine learning approach
integrating oversampling and adaptive boosting techniques. AIMS Med. Sci. 11(2), 58–71 (2024).
57. Guido, R. et al. An overview on the advancements of support vector machine models in healthcare applications: A review.
Information 15(4), 235. https://doi.org/10.3390/info15040235 (2024).
58. El Hamdaoui, H. et al. Improving heart disease prediction using random forest and adaboost algorithms. iJOE 17(11), 61 (2021).
59. Wassan, S. et al. Gradient boosting for health IoT federated learning. Sustainability 14, 16842. https://doi.org/10.3390/su142416842
(2022).
60. Li, W., Peng, Y. & Peng, K. Diabetes prediction model based on GA-XGBoost and stacking ensemble algorithm. PLOS ONE 19(9),
e0311222 (2024).
61. Prasetyo, S. Y. & Izdihar, Z. N. Multi-layer perceptron approach for diabetes risk prediction using BRFSS data. In 2024 IEEE 10th
International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA). (2024).
62. Aria, M., et al., Acute lymphoblastic leukemia (ALL) image dataset. Kaggle, (2021).
63. Aria, M., et al., COVID-19 Lung CT scans: A large dataset of lung CT scans for COVID-19 (SARS-CoV-2) detection. Kaggle. https:​​​
//w​ww.kag​gle​.com/mehra​daria/c​ovid​19​​-lung-ct-scans, accessed 20 April 2021, (2021).
64. Aria, M., Hashemzadeh, M. & Farajzadeh, N. QDL-CMFD: A quality-independent and deep learning-based copy-move image
forgery detection method. Neurocomputing 511, 213–236 (2022).
65. Farhad, A. et al. Artificial intelligence in estimating fractional flow reserve: A systematic literature review of techniques. BMC
Cardiovas. Disord. 23(1), 407 (2023).
66. Aria, M., Nourani, E. & Golzari Oskouei, A. ADA-COVID: Adversarial deep domain adaptation-based diagnosis of COVID-19
from lung CT scans using triplet embeddings. Comput. Intell. Neurosci. 2022(1), 2564022 (2022).

Author contributions
S.Z.H., R.R., H.E. and M.A. were responsible for the conceptualization and design of the study. S.Z.H and M.A.
performed the development and evaluation of the ML and DL models. M.K., M.A, and V.Z. prepared the datasets
and also facilitated the process of data gathering. S.Z.H. handled data gathering from the EHR and Iran Health
Insurance System, as well as telephone interviews with patients and data cleansing. The initial draft was critically
reviewed by R.R. and M.A. and ultimately all authors read and approved the final version of the manuscript.

Funding
Throughout this study, no financial resources or funding were received.

Declarations

Conflict of interest
The authors declare no competing interests.

Ethical approval
All experimental protocols were approved by the Institutional Review Board of Shahid Beheshti University
of Medical Sciences, with the approval code IR.SBMU.RETECH.REC.1401.823, and informed consent was
obtained from all subjects and/or their legal guardians. In addition, all methods were performed in accordance
with the relevant guidelines and regulations.

Additional information
Supplementary Information The online version contains supplementary material available at ​h​t​t​p​s​:​/​/​d​o​i.​ ​o​r​g​/​1​
0​.​1​03​ ​8​/​s​4​1​5​9​8​-0​ ​2​4​-​8​1​7​3​4​-​y​​​​.​ ​​
Correspondence and requests for materials should be addressed to R.R. or M.A.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 17

www.nature.com/scientificreports/

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives

4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in
any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide
a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have
permission under this licence to share adapted material derived from this article or parts of it. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence
and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to
obtain permission directly from the copyright holder. To view a copy of this licence, visit ​h​t​t​p​:​//​ ​c​r​e​a​t​i​v​e​co
​ ​m​m​o​
n​s​.​o​r​g/​ ​l​i​c​e​n​s​e​s​/b
​ ​y​-​n​c​-​n​d​/​4​.​0​/​​​​.​ ​​

© The Author(s) 2024

Scientific Reports | (2024) 14:30147 | https://doi.org/10.1038/s41598-024-81734-y 18