Devops based progressive delivery finalized
- 1. DevOps Based Progressive Delivery Telehealth Platform Bhagvan Kommadi DIRECTOR OF PRODUCT ENGINEERING, VALUEMOMENTUM LEADERSHIP
- 2. © 2021 All Rights Reserved. 2 Mind Map - DevOps Telehealth Solution Architectural Components Deep Learning Solution Implementation What’s Next ? Sneak Preview of Upcoming Features Agenda DevOps Agility Flexibility Quality
- 3. © 2021 All Rights Reserved. 3 3 Mind map No Code Platforms Low Code Platforms Market Requirements Health Insurance Payment Processing AI Digital health Data Sources Knowledge Resources Legacy Systems & Migration Budget In House Applications Health Care Application Upgrades & Update Training Metadata Customizations Licensing & Subscription Life insurance Tech Stack Wellness Apps Hospital Management Application Scope Neural Sketch Learning AI/ML Devops Containers Cloud CI/CD Version Control Testing GIT IDE Automation Virtualization Network Monitoring System Admin
- 5. © 2021 All Rights Reserved. 5 Patient Journey Book an Appointment Appointment Reminder Schedule Tracking Intimate Delays Manage Appointments Post Appointment Engagement
- 6. © 2021 All Rights Reserved. 6 What Patients Want & Doctor’s View
- 7. © 2021 All Rights Reserved. 7 Voice Assistants
- 8. © 2021 All Rights Reserved. 8 Process Flow Filters NLP API Web Mobile Desktop Queries Registry Data Sources Links Knowledge Base Updates 1) Diseases 2) Data Sources 3) Artifacts Top Down As Knowledge Domain Expert Bottom Up Registry & knowledge updates • Orphanet • Johns Hopkins • WHO • NORD
- 9. © 2021 All Rights Reserved. 9 Knowledgebase Architecture Service Layer Visualization Layer Data Access Layer Knowledg e Updates Internet Data Sources Search Manager Knowledge Base Manager axios cheerios node Disease Registry Disease Knowledge
- 10. © 2021 All Rights Reserved. 10 Digital Assistants & Video Bots Taking clinical notes Retrieving cases Updating Prescriptions Creating consultation appointments
- 12. © 2021 All Rights Reserved. 12 Bug Tracking Systems Monitoring Systems Configuration Management OS Patching CI/CD Pipelines Containers Source Code Control Systems Collaboration Infrastructure Foundational Tools & Infrastructure
- 13. © 2021 All Rights Reserved. 13 DevOps Evolution - Metrics Metric Value Deployment Frequency Monthly Lead Time Between a week and 6months MTTR Less than a week Change Failure Rate Less than 15% Low Mid High Metric Value Deployment Frequency Between Daily and Weekly Lead Time Less than a week MTTR Less than a day Change Failure Rate Less than 15% Metric Value Deployment Frequency On Demand Lead Time Less than an hour MTTR Less than an hour Change Failure Rate Less than 5%
- 14. © 2021 All Rights Reserved. 14 DevOps Evolution - Challenges • Resistance to Change • Legacy Architecture • Shortage of Skills • Lack of Automation • Unclear Objectives Low • Resistance to Change • Legacy Architecture • Shortage of Skills • Lack of Automation Mid • Legacy Architecture • Shortage of Skills High
- 15. © 2021 All Rights Reserved. 15 Flow of Change Business Domain Business Domain Stream Aligned Team Stream Aligned Team Enabling Team Platform Team Complex Subsystem Team Flow of Change Collaboration X as a Service Facilitating
- 16. © 2021 All Rights Reserved. 16 Jenkins Devops - Features Authorization Administration Analytics Scalability Availability Distributed Teams Container Support Quality Cost Product Time
- 17. © 2021 All Rights Reserved. 17 Jenkins Pipeline – Devops Workflow Code Checkout Build the code Test the code Stage the build Release the build Deploy the build Version Control Continuous Integration Continuous Delivery Continuous Deployment
- 18. © 2021 All Rights Reserved. 18 Progressive Delivery of Web Apps Automating the Delivery of Web Apps Jenkins Server GitRepo Update Jenkins Trigger Jenkins Build Process AWS Deployment Live Web App
- 19. © 2021 All Rights Reserved. 19 Progressive Delivery of Mobile Apps Automating the Delivery of Mobile and Web Apps Jenkins Server GitRepo Update Jenkins Trigger Jenkins Build Process Google Play Store Apple App Store
- 20. © 2021 All Rights Reserved. 20 Jenkins Devops : Benefits Autonomous Teams Simple setup Delivers Release in Time Faster Delivery Higher Productivity Software Predictivity Business Sustainability Autonomous Teams Audit Trails End to End Integration Tests Functional Tests Unit Tests
- 21. © 2021 All Rights Reserved. 21 Jenkins – Agent based Architecture Jenkins Server Git Lab Jenkins Agent Jenkins Agent Jenkins Agent
- 22. © 2021 All Rights Reserved. 22 Dev Ops Best Practices 22 Automating Software Delivery Treating Release Orchestration as a Business Imperative Adopting Best-of-Breed Solutions but Doing so Judiciously Practicing Value Stream Thinking Using Feature Flags for Safer, Faster Software Delivery
- 23. Deep Learning Solution HealthCare Data Analysis
- 24. © 2021 All Rights Reserved. 24 Input Layer Hidden Layer Hidden Layer Output Layer Hidden Layer Hidden Layer Convolutional Neural Network
- 25. © 2021 All Rights Reserved. 25 Image Pixels Pixel Value Color Depth Grey Scale Intensity CNN Layers Convolution Rectified linear unit (ReLu) Pooling Fully Connected Features Extraction Image Recognition
- 26. © 2021 All Rights Reserved. 26 CNN Models Keras pre- trained CNNs Xception VGG16 VGG1 9 ResNet 50 Inception V3 Inception ResNetV2 Mobile Net Dense Net NASNet MobileNet V2
- 27. © 2021 All Rights Reserved. 27 Atelectasis 192 Cardiomegaly 50 Consolidation 72 Edema 41 Effusion 203 Emphysema 42 Fibrosis 38 Hernia 5 Infiltration 503 Mass 99 No Finding 3044 Nodule 144 Pleural_Thickening 65 Pneumonia 14 Pneumothorax 114 Total Result 5606 Atelectasis Pneumothorax Gender : Male, Female Xray Position: Frontal , Back, Side Lung Disease Analysis
- 28. © 2021 All Rights Reserved. 28 Training Testing Validation Normal Infection by Virus Infection by Bacteria True Positive False Positive False Negative False Positive Actual P r e d i c t e d Precision = True Positive/Actual Results Recall = True Positive/Predicted Results Accuracy = (True Positive + True Negative)/Total Actual Results = True Positive + False Positive Predicted Results = True Positive + False Negative CNN Scale Shift Rotate Salt and Pepper noise Flip Lung Disease – Deep Learning
- 29. © 2021 All Rights Reserved. 29 Total params: 12,167,424 Trainable params: 12,167,424 Non-trainable params: 0 Precision: 0.734694 % Recall: 0.073022 % Fscore: 0.261248 % Training Data Bias = Average Predicted Value - Average Actual Value Convolutional Neural Network Results
- 30. © 2021 All Rights Reserved. 30 Convolutional Neural Network - Basic
- 31. © 2021 All Rights Reserved. 31 Convolutional Neural Network – Complete Dataset
- 33. © 2021 All Rights Reserved. 33 White Coats Platform Network App Practice Plus Clinic Plus StaffPlus Patient Plus Registered Doctors* 3,00,000+ Locations 3,000+ Onboarded Doctors (App) 50,000+ Doctors from Metro/Tier 1 & 2 ~68% Doctors from Tier 3 & 4 ~32% # of Specialties 125+ General Practitioners ~35% % of Drs : Specialists ~65% • App & Web Views Q1- 2.05L ; Q2- 2.8L (Growth 36%) • Social Interactions (APP) Q1- 1.27L ; Q2- 1.58L (Growth 25%) • Network Cases Q1- 800 ; Q2- 1000 (Growth 25%) White Coats Platform
- 34. © 2021 All Rights Reserved. 34 What’s Next ? Relation of programming paradigms to CD and deployment experience Differences between prevalence and effects of intentional vs. unintentional violations of SOLID OO design principles Count of manual interventions required to go from dev to production and relation to software architecture and testing strategy Relation of database normalization to deployment and incident/rollback frequency Complexity of DevOps pipeline in terms of tool count Overall effect of CD on application quality and security Prevalence of version control for various definable (source, build scripts, infrastructure configuration, etc.)
- 35. © 2021 All Rights Reserved. 35 Moving From DevOps to -> DevOps Data Ops Dev SecOps AI Ops MLOps Development Operati ons Security Data DevOps DevSecOps DataOps
- 36. © 2021 All Rights Reserved. 36 Ops Roles - Future Vendor Engg Product Engg SocioTechnical Systems Engg Portfolio Management
- 37. © 2021 All Rights Reserved. 37 Clinical Knowledge - EBM 37 • Guidelines • Decisions • Study Results • Observations Data Evidence Guidance Action Omni Channel Telehealth Platform Genomic Applications Personalized Medicine Evidence Based Medicine Predictive Toxicology QSAR Chemo-Informatics
- 38. Thank you! Bhagvan Kommadi DIRECTOR OF PRODUCT ENGG, VALUEMOMENTUM
Editor's Notes
- #2: This PowerPoint Template includes a series of slide masters with predefined layouts and color schemes for formatting slides Slide Masters are displayed when you right click on a slide and select Layout from menu
- #6: Book consultations online (Video/Audio) Get medicines delivered Book diagnosis tests Set reminders for medicines Ask questions to experts Add and save medical records View appointments and diagnosis history View chat history and order history Add and view family members View and update profile Read articles and share them
- #17: An open-source CI server that enables development and operation teams to reliably build, test, and deploy the software with shared pipeline and error tracking function. It can also be associated with a variety of dashboard plugins.
- #18: Continuous Development: Git, SVN, Mercurial, CVS, Jira Continuous Integration: Jenkins, Bamboo, Hudson Continuous Delivery: Nexus, Archiva, Tomcat Continuous Deployment: Puppet, Chef, Docker Continuous Monitoring: Splunk, ELK Stack, Nagios Continuous Testing: Selenium, Katalon Studio
- #21: Merge the project’s source code into a single repository and maintain. Automate the build and integration. Have the build self-test. Regularly update changes to the baseline Add all commits to the baseline Accelerate the builds Run tests in a clone of the production environment Make it easy to get the latest deliverables Build results can be monitored easily to public Automate deployment
- #24: Image consisting of pixels and greyscale intensity ranges from 0 to 255 and weighted average is measured by the Perceptron
- #35: we defined the SOLID principles as follows: • Single-responsibility principle – Every class should have only one reason to change. • Open-closed principle – Every class should be open for extension but closed for modification. • Liskov substitution principle – Every subclass should be substitutable for its base class. • Interface segregation principle – No interface should require a client to worry about things it does not want from the implementation. • Dependency inversion principle – No abstraction should depend on concrete implementation details.
- #36: DevOps is a union in which processes, products, and people are integrated to provide Continuous Integration and Continuous Delivery of value to end-users.