msk_admin/Lumina-MSK
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update the architecture & description #1

Merged
msk_admin merged 1 commits from update_ into main 2026-06-24 03:35:20 +00:00
Conversation 1 Commits 1 Files Changed 77 +8999 -14
DTJ-Tran commented 2026-06-24 03:34:44 +00:00
Contributor
Copy Link

Summary by CodeRabbit

Release Notes

  • New Features

    • Launched web interface for image analysis supporting drag-and-drop file upload, multi-model selection, and segmentation overlay visualization.
    • Enabled PDF report generation with patient data management, save, and export functionality.

  • Documentation

    • Added detailed architecture specifications covering system design and deployment topology.
    • Updated infrastructure component documentation (pgvector embeddings storage, cloud/browser LLM integration).
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit ## Release Notes * **New Features** * Launched web interface for image analysis supporting drag-and-drop file upload, multi-model selection, and segmentation overlay visualization. * Enabled PDF report generation with patient data management, save, and export functionality. * **Documentation** * Added detailed architecture specifications covering system design and deployment topology. * Updated infrastructure component documentation (pgvector embeddings storage, cloud/browser LLM integration). <!-- end of auto-generated comment: release notes by coderabbit.ai -->
coderabbitai[bot] commented 2026-06-24 03:35:00 +00:00 (Migrated from github.com)
Copy Link

Review Change Stack

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info
⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 174d111b-6b35-41da-8481-a249cde088e0

📥 Commits

Reviewing files that changed from the base of the PR and between 16a91bd17e and f705113711.

⛔ Files ignored due to path filters (7)
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/Roboto-Bold.ttf is excluded by !**/*.ttf
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/Roboto-Italic.ttf is excluded by !**/*.ttf
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/Roboto-Regular.ttf is excluded by !**/*.ttf
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/arial.ttf is excluded by !**/*.ttf
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/arialbd.ttf is excluded by !**/*.ttf
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/ariali.ttf is excluded by !**/*.ttf
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/logo.png is excluded by !**/*.png
📒 Files selected for processing (70)
  • README.md
  • proj_level_reading/ARCHITECT/SOFTWARE_ARCHITECTURE_SPEC.md
  • proj_level_reading/ARCHITECT/SOLUTION_ARCHITECTURE_SPEC.md
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/Dockerfile
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/app.py
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/pdf_service.py
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/templates/css/style.css
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/templates/index.html
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/templates/js/script.js
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ML/requirements.txt
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/README.md
  • workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/architecture-review.html
  • workspace/LEGACY/VKIST_ML/docs/Instruction_Manunal_API_Doc_Setup_pilot.md
  • workspace/LEGACY/VKIST_ML/docs/PILOT_TECH_DESC.md
  • workspace/LEGACY/VKIST_ML/docs/VKIST_ULTRASOUND_INTRODUCE_SPEC.md
  • workspace/LEGACY/VKIST_ML/docs/arch_code_analysis.md
  • workspace/sprint_1_2/CODEBASE/backend/spec/backend-spec.md
  • workspace/sprint_1_2/CODEBASE/backend/spec/interface-contract.md
  • workspace/sprint_1_2/CODEBASE/data/spec/data_spec.md
  • workspace/sprint_1_2/CODEBASE/data/spec/interface_contract.md
  • workspace/sprint_1_2/CODEBASE/frontend/spec/frontend_spec.md
  • workspace/sprint_1_2/CODEBASE/frontend/spec/interface_contract.md
  • workspace/sprint_1_2/CODEBASE/frontend/subprojects/offline/spec/interface_contract.md
  • workspace/sprint_1_2/CODEBASE/frontend/subprojects/offline/spec/offline_spec.md
  • workspace/sprint_1_2/CODEBASE/infra/implementation/cicd/jenkin_run.sh
  • workspace/sprint_1_2/CODEBASE/infra/implementation/observability/docker-compose.yaml
  • workspace/sprint_1_2/CODEBASE/infra/implementation/observability/prometheus.yml
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_convnext_tiny/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_densenet/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_efficientnet/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_resnet50/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_swin_v2_s/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/generate_ensemble.py
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/inflammation_model_efficientnet_b0_ultrasound_2_cls/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/msk_vision_pipeline_ensemble/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_post_deeplabv3/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_post_deeplabv3_resnet101/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_post_efficientfeedback/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_unet3plus_att/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_unet_resnet101/config.pbtxt
  • workspace/sprint_1_2/CODEBASE/infra/implementation/s3/upload_ensemble.py
  • workspace/sprint_1_2/CODEBASE/infra/implementation/sh_files/mv_file.sh
  • workspace/sprint_1_2/CODEBASE/infra/implementation/sh_files/mv_sh_2.sh
  • workspace/sprint_1_2/CODEBASE/infra/implementation/sh_files/mv_sh_3.sh
  • workspace/sprint_1_2/CODEBASE/infra/implementation/sh_files/mv_sh_4.sh
  • workspace/sprint_1_2/CODEBASE/infra/implementation/triton_run/modal_triton.py
  • workspace/sprint_1_2/CODEBASE/infra/implementation/triton_run/run.sh
  • workspace/sprint_1_2/CODEBASE/infra/spec/infra_spec.md
  • workspace/sprint_1_2/CODEBASE/infra/spec/interface_contract.md
  • workspace/sprint_1_2/CODEBASE/knowledge/spec/interface_contract.md
  • workspace/sprint_1_2/CODEBASE/knowledge/spec/knowledge_spec.md
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/efficientfeedback.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/__init__.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/automatic_mask_generator.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/build_sam.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/__init__.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/common.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/image_encoder.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/mask_decoder.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/prompt_encoder.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/sam.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/transformer.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/predictor.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/utils/__init__.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/utils/amg.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/utils/onnx.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/utils/transforms.py
  • workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/unet3plus_att.py
  • workspace/sprint_1_2/Design_Material/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy
  • workspace/sprint_1_2/SPRINT_1_2_ARCHITECTURE_SPEC.md
📝 Walkthrough

Walkthrough

The PR adds two project-level architecture specification documents (solution and software architecture) under proj_level_reading/ARCHITECT/, a new sprint 1–2 architecture spec, and minor updates to backend/knowledge component specs replacing Qdrant with pgvector and PhoGPT with GemmaE2B. It also introduces a complete legacy VKIST NON_ML ultrasound FastAPI application including ML inference pipeline, PDF report generation, Vietnamese web UI, Dockerfile, and supporting documentation.

Changes

Architecture & Design Documentation

Layer / File(s) Summary
Project-level solution and software architecture specs
proj_level_reading/ARCHITECT/SOLUTION_ARCHITECTURE_SPEC.md, proj_level_reading/ARCHITECT/SOFTWARE_ARCHITECTURE_SPEC.md		 Adds two complete architecture specification documents covering the hybrid edge/on-prem platform: NFR-16/NFR-16a air-gap constraints, inference fallback chain (WebLLM → Triton → Vertex AI → deterministic templates), RAG-Referee safety gates, Decree 13 PII redaction pipelines, Redis/Postgres/Triton gRPC contracts, C4 context/container/component/deployment diagrams, REST/SSE contracts, guardrail worker topology, build-vs-buy matrix, phased task decomposition, and execution timeline.
Sprint 1-2 architecture spec and component spec updates
workspace/sprint_1_2/SPRINT_1_2_ARCHITECTURE_SPEC.md, workspace/sprint_1_2/CODEBASE/backend/spec/backend-spec.md, workspace/sprint_1_2/CODEBASE/knowledge/spec/knowledge_spec.md, README.md		 Adds a full sprint-scoped architecture spec with C4 PlantUML diagrams, ML workflow sequence, NFR coverage, compliance obligations, infrastructure decisions, and open questions. Updates backend-spec.md and knowledge_spec.md to replace Qdrant with Postgres pgvector, replace PhoGPT/MedGPT with GemmaE2B/MedGemma, and add mandatory RAG pre-generation retrieval. Updates README directory tree to reflect PROJ_LEVEL_READING/ and Design_Material/ additions.

Legacy VKIST NON_ML Ultrasound Application

Layer / File(s) Summary
FastAPI ML inference pipeline
workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/app.py (lines 1–560)		 Adds global config, device selection, label/color constants, model-loading helpers for angle/inflammation/SUP/POST segmentation, preprocessing transforms, inference utilities, mask geometry helpers (get_mask_bounding_box, find_max_continuous_segment), thickness measurement (measure_thickness_new), inflammation severity scoring, and visualization utilities (create_segmentation_overlay, apply_clahe).
FastAPI API endpoints and PDF report service
workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/app.py (lines 562–852), workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/pdf_service.py		 Adds POST /api/analyze orchestrating the full ML pipeline, GET /api/health, POST /api/save (timestamped patient folder with info.txt, images, and auto-generated PDF), POST /api/export-pdf returning PDF as attachment, and Uvicorn launcher. Adds pdf_service.py with MedicalReportPDF (FPDF subclass), get_clean_image_stream (base64/path to non-interlaced PNG), and generate_medical_report assembling the full Vietnamese-language patient report.
Vietnamese web UI
workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/templates/index.html, ...templates/css/style.css, ...templates/js/script.js		 Adds index.html with model-selection dropdowns, drag-drop upload panel, results cards (angle/inflammation/segmentation/measurement/severity), patient info form, and diagnosis modal. Adds style.css with full layout grid, form/badge/modal/spinner styling and responsive breakpoint. Adds script.js with file handling, uploadAndAnalyze, displayResults, save/export wiring to /api/save and /api/export-pdf, color legend rendering, modal lifecycle, and page-load health check.
Dockerfile, requirements, README, and architecture review
workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/Dockerfile, .../ML/requirements.txt, .../README.md, .../architecture-review.html		 Adds a multi-stage Dockerfile (python:3.12 builder + python:3.12-slim runtime). Populates requirements.txt with pinned PyTorch CPU and FastAPI/CV/PDF dependencies. Adds a README.md with setup, model-weight instructions, and LEGACY repo rules. Adds a static architecture-review.html proposing app.py modularization with before/after Mermaid diagrams.

Sequence Diagram(s)

sequenceDiagram
    rect rgba(173, 216, 230, 0.5)
        Note over Browser,FastAPI: Image Upload & Analysis
    end
    participant Browser
    participant FastAPI
    participant AngleModel
    participant SegmentationModel
    participant PDFService
    participant MinIO

    Browser->>FastAPI: POST /api/analyze (image file + model params)
    FastAPI->>FastAPI: apply_clahe → base64 enhanced image
    FastAPI->>AngleModel: predict_angle(image)
    AngleModel-->>FastAPI: angle_class, confidence
    FastAPI->>FastAPI: predict_inflammation(image)
    FastAPI->>SegmentationModel: segment_image(image, angle_type)
    SegmentationModel-->>FastAPI: masks
    FastAPI->>FastAPI: measure_thickness_new(masks) + analyze_inflammation_severity(masks)
    FastAPI->>FastAPI: create_segmentation_overlay(image, masks, measurement)
    FastAPI-->>Browser: JSON (enhanced_img, segmented_img, metadata)

    rect rgba(144, 238, 144, 0.5)
        Note over Browser,MinIO: Save & PDF Export
    end
    Browser->>FastAPI: POST /api/save (patient_info + analysis_result + images)
    FastAPI->>FastAPI: create timestamped patient folder, write info.txt, save images
    FastAPI->>PDFService: generate_medical_report(patient_info, analysis_result, images)
    PDFService-->>FastAPI: PDF bytes
    FastAPI->>MinIO: store report artifact
    FastAPI-->>Browser: { saved_dir }

    Browser->>FastAPI: POST /api/export-pdf (same payload)
    FastAPI->>PDFService: generate_medical_report(...)
    PDFService-->>FastAPI: PDF bytes
    FastAPI-->>Browser: PDF attachment download

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~60 minutes

Poem

🐇 Hippity-hop, the specs have landed,
C4 diagrams neatly branded!
RAG-Referee guards every token,
Air-gap rules shall not be broken.
Legacy app.py, segmenting knees,
PDF reports typed with ease~
This bunny approves — proceed, if you please! 🦺

✨ Finishing Touches
📝 Generate docstrings
  • Create stacked PR
  • Commit on current branch
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Commit unit tests in branch update_

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

<!-- This is an auto-generated comment: summarize by coderabbit.ai --> <!-- review_stack_entry_start --> [![Review Change Stack](https://storage.googleapis.com/coderabbit_public_assets/review-stack-in-coderabbit-ui.svg)](https://app.coderabbit.ai/change-stack/DTJ-Tran/pilot_msk_ultrasound_stack/pull/1?utm_source=github_walkthrough&utm_medium=github&utm_campaign=change_stack) <!-- review_stack_entry_end --> <!-- This is an auto-generated comment: failure by coderabbit.ai --> > [!CAUTION] > ## Review failed > > The pull request is closed. <!-- end of auto-generated comment: failure by coderabbit.ai --> <details> <summary>ℹ️ Recent review info</summary> <details> <summary>⚙️ Run configuration</summary> **Configuration used**: defaults **Review profile**: CHILL **Plan**: Pro **Run ID**: `174d111b-6b35-41da-8481-a249cde088e0` </details> <details> <summary>📥 Commits</summary> Reviewing files that changed from the base of the PR and between 16a91bd17ee7dfbf7922c59ac6e4e57ae664d734 and f70511371126afcc47c62a94cc85d4899214520e. </details> <details> <summary>⛔ Files ignored due to path filters (7)</summary> * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/Roboto-Bold.ttf` is excluded by `!**/*.ttf` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/Roboto-Italic.ttf` is excluded by `!**/*.ttf` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/Roboto-Regular.ttf` is excluded by `!**/*.ttf` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/arial.ttf` is excluded by `!**/*.ttf` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/arialbd.ttf` is excluded by `!**/*.ttf` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/fonts/ariali.ttf` is excluded by `!**/*.ttf` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/assets/logo.png` is excluded by `!**/*.png` </details> <details> <summary>📒 Files selected for processing (70)</summary> * `README.md` * `proj_level_reading/ARCHITECT/SOFTWARE_ARCHITECTURE_SPEC.md` * `proj_level_reading/ARCHITECT/SOLUTION_ARCHITECTURE_SPEC.md` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/Dockerfile` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/app.py` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/pdf_service.py` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/templates/css/style.css` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/templates/index.html` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/templates/js/script.js` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ML/requirements.txt` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/README.md` * `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/architecture-review.html` * `workspace/LEGACY/VKIST_ML/docs/Instruction_Manunal_API_Doc_Setup_pilot.md` * `workspace/LEGACY/VKIST_ML/docs/PILOT_TECH_DESC.md` * `workspace/LEGACY/VKIST_ML/docs/VKIST_ULTRASOUND_INTRODUCE_SPEC.md` * `workspace/LEGACY/VKIST_ML/docs/arch_code_analysis.md` * `workspace/sprint_1_2/CODEBASE/backend/spec/backend-spec.md` * `workspace/sprint_1_2/CODEBASE/backend/spec/interface-contract.md` * `workspace/sprint_1_2/CODEBASE/data/spec/data_spec.md` * `workspace/sprint_1_2/CODEBASE/data/spec/interface_contract.md` * `workspace/sprint_1_2/CODEBASE/frontend/spec/frontend_spec.md` * `workspace/sprint_1_2/CODEBASE/frontend/spec/interface_contract.md` * `workspace/sprint_1_2/CODEBASE/frontend/subprojects/offline/spec/interface_contract.md` * `workspace/sprint_1_2/CODEBASE/frontend/subprojects/offline/spec/offline_spec.md` * `workspace/sprint_1_2/CODEBASE/infra/implementation/cicd/jenkin_run.sh` * `workspace/sprint_1_2/CODEBASE/infra/implementation/observability/docker-compose.yaml` * `workspace/sprint_1_2/CODEBASE/infra/implementation/observability/prometheus.yml` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_convnext_tiny/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_densenet/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_efficientnet/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_resnet50/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/angle_classify_swin_v2_s/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/generate_ensemble.py` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/inflammation_model_efficientnet_b0_ultrasound_2_cls/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/msk_vision_pipeline_ensemble/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_post_deeplabv3/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_post_deeplabv3_resnet101/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_post_efficientfeedback/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_unet3plus_att/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/segmentation_model_unet_resnet101/config.pbtxt` * `workspace/sprint_1_2/CODEBASE/infra/implementation/s3/upload_ensemble.py` * `workspace/sprint_1_2/CODEBASE/infra/implementation/sh_files/mv_file.sh` * `workspace/sprint_1_2/CODEBASE/infra/implementation/sh_files/mv_sh_2.sh` * `workspace/sprint_1_2/CODEBASE/infra/implementation/sh_files/mv_sh_3.sh` * `workspace/sprint_1_2/CODEBASE/infra/implementation/sh_files/mv_sh_4.sh` * `workspace/sprint_1_2/CODEBASE/infra/implementation/triton_run/modal_triton.py` * `workspace/sprint_1_2/CODEBASE/infra/implementation/triton_run/run.sh` * `workspace/sprint_1_2/CODEBASE/infra/spec/infra_spec.md` * `workspace/sprint_1_2/CODEBASE/infra/spec/interface_contract.md` * `workspace/sprint_1_2/CODEBASE/knowledge/spec/interface_contract.md` * `workspace/sprint_1_2/CODEBASE/knowledge/spec/knowledge_spec.md` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/efficientfeedback.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/__init__.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/automatic_mask_generator.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/build_sam.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/__init__.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/common.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/image_encoder.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/mask_decoder.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/prompt_encoder.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/sam.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/modeling/transformer.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/predictor.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/utils/__init__.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/utils/amg.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/utils/onnx.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/segment_anything/utils/transforms.py` * `workspace/sprint_1_2/CODEBASE/ml/implementation/cv/arch/unet3plus_att.py` * `workspace/sprint_1_2/Design_Material/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy` * `workspace/sprint_1_2/SPRINT_1_2_ARCHITECTURE_SPEC.md` </details> </details> --- <!-- walkthrough_start --> <details> <summary>📝 Walkthrough</summary> ## Walkthrough The PR adds two project-level architecture specification documents (solution and software architecture) under `proj_level_reading/ARCHITECT/`, a new sprint 1–2 architecture spec, and minor updates to backend/knowledge component specs replacing Qdrant with pgvector and PhoGPT with GemmaE2B. It also introduces a complete legacy VKIST NON_ML ultrasound FastAPI application including ML inference pipeline, PDF report generation, Vietnamese web UI, Dockerfile, and supporting documentation. ## Changes **Architecture & Design Documentation** |Layer / File(s)|Summary| |---|---| |**Project-level solution and software architecture specs** <br> `proj_level_reading/ARCHITECT/SOLUTION_ARCHITECTURE_SPEC.md`, `proj_level_reading/ARCHITECT/SOFTWARE_ARCHITECTURE_SPEC.md`|Adds two complete architecture specification documents covering the hybrid edge/on-prem platform: NFR-16/NFR-16a air-gap constraints, inference fallback chain (WebLLM → Triton → Vertex AI → deterministic templates), RAG-Referee safety gates, Decree 13 PII redaction pipelines, Redis/Postgres/Triton gRPC contracts, C4 context/container/component/deployment diagrams, REST/SSE contracts, guardrail worker topology, build-vs-buy matrix, phased task decomposition, and execution timeline.| |**Sprint 1-2 architecture spec and component spec updates** <br> `workspace/sprint_1_2/SPRINT_1_2_ARCHITECTURE_SPEC.md`, `workspace/sprint_1_2/CODEBASE/backend/spec/backend-spec.md`, `workspace/sprint_1_2/CODEBASE/knowledge/spec/knowledge_spec.md`, `README.md`|Adds a full sprint-scoped architecture spec with C4 PlantUML diagrams, ML workflow sequence, NFR coverage, compliance obligations, infrastructure decisions, and open questions. Updates backend-spec.md and knowledge_spec.md to replace Qdrant with Postgres pgvector, replace PhoGPT/MedGPT with GemmaE2B/MedGemma, and add mandatory RAG pre-generation retrieval. Updates README directory tree to reflect `PROJ_LEVEL_READING/` and `Design_Material/` additions.| **Legacy VKIST NON_ML Ultrasound Application** |Layer / File(s)|Summary| |---|---| |**FastAPI ML inference pipeline** <br> `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/app.py` (lines 1–560)|Adds global config, device selection, label/color constants, model-loading helpers for angle/inflammation/SUP/POST segmentation, preprocessing transforms, inference utilities, mask geometry helpers (`get_mask_bounding_box`, `find_max_continuous_segment`), thickness measurement (`measure_thickness_new`), inflammation severity scoring, and visualization utilities (`create_segmentation_overlay`, `apply_clahe`).| |**FastAPI API endpoints and PDF report service** <br> `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/app.py` (lines 562–852), `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/pdf_service.py`|Adds `POST /api/analyze` orchestrating the full ML pipeline, `GET /api/health`, `POST /api/save` (timestamped patient folder with info.txt, images, and auto-generated PDF), `POST /api/export-pdf` returning PDF as attachment, and Uvicorn launcher. Adds `pdf_service.py` with `MedicalReportPDF` (FPDF subclass), `get_clean_image_stream` (base64/path to non-interlaced PNG), and `generate_medical_report` assembling the full Vietnamese-language patient report.| |**Vietnamese web UI** <br> `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/templates/index.html`, `...templates/css/style.css`, `...templates/js/script.js`|Adds `index.html` with model-selection dropdowns, drag-drop upload panel, results cards (angle/inflammation/segmentation/measurement/severity), patient info form, and diagnosis modal. Adds `style.css` with full layout grid, form/badge/modal/spinner styling and responsive breakpoint. Adds `script.js` with file handling, `uploadAndAnalyze`, `displayResults`, save/export wiring to `/api/save` and `/api/export-pdf`, color legend rendering, modal lifecycle, and page-load health check.| |**Dockerfile, requirements, README, and architecture review** <br> `workspace/LEGACY/VKIST_ML/codebase-vkist-ultrasound-legacy/ NON_ML/Dockerfile`, `.../ML/requirements.txt`, `.../README.md`, `.../architecture-review.html`|Adds a multi-stage Dockerfile (python:3.12 builder + python:3.12-slim runtime). Populates `requirements.txt` with pinned PyTorch CPU and FastAPI/CV/PDF dependencies. Adds a `README.md` with setup, model-weight instructions, and LEGACY repo rules. Adds a static `architecture-review.html` proposing `app.py` modularization with before/after Mermaid diagrams.| ## Sequence Diagram(s) ```mermaid sequenceDiagram rect rgba(173, 216, 230, 0.5) Note over Browser,FastAPI: Image Upload & Analysis end participant Browser participant FastAPI participant AngleModel participant SegmentationModel participant PDFService participant MinIO Browser->>FastAPI: POST /api/analyze (image file + model params) FastAPI->>FastAPI: apply_clahe → base64 enhanced image FastAPI->>AngleModel: predict_angle(image) AngleModel-->>FastAPI: angle_class, confidence FastAPI->>FastAPI: predict_inflammation(image) FastAPI->>SegmentationModel: segment_image(image, angle_type) SegmentationModel-->>FastAPI: masks FastAPI->>FastAPI: measure_thickness_new(masks) + analyze_inflammation_severity(masks) FastAPI->>FastAPI: create_segmentation_overlay(image, masks, measurement) FastAPI-->>Browser: JSON (enhanced_img, segmented_img, metadata) rect rgba(144, 238, 144, 0.5) Note over Browser,MinIO: Save & PDF Export end Browser->>FastAPI: POST /api/save (patient_info + analysis_result + images) FastAPI->>FastAPI: create timestamped patient folder, write info.txt, save images FastAPI->>PDFService: generate_medical_report(patient_info, analysis_result, images) PDFService-->>FastAPI: PDF bytes FastAPI->>MinIO: store report artifact FastAPI-->>Browser: { saved_dir } Browser->>FastAPI: POST /api/export-pdf (same payload) FastAPI->>PDFService: generate_medical_report(...) PDFService-->>FastAPI: PDF bytes FastAPI-->>Browser: PDF attachment download ``` ## Estimated code review effort 🎯 4 (Complex) | ⏱️ ~60 minutes ## Poem > 🐇 Hippity-hop, the specs have landed, > C4 diagrams neatly branded! > RAG-Referee guards every token, > Air-gap rules shall not be broken. > Legacy app.py, segmenting knees, > PDF reports typed with ease~ > *This bunny approves — proceed, if you please!* 🦺 </details> <!-- walkthrough_end --> <!-- finishing_touch_checkbox_start --> <details> <summary>✨ Finishing Touches</summary> <details> <summary>📝 Generate docstrings</summary> - [ ] <!-- {"checkboxId": "7962f53c-55bc-4827-bfbf-6a18da830691"} --> Create stacked PR - [ ] <!-- {"checkboxId": "3e1879ae-f29b-4d0d-8e06-d12b7ba33d98"} --> Commit on current branch </details> <details> <summary>🧪 Generate unit tests (beta)</summary> - [ ] <!-- {"checkboxId": "f47ac10b-58cc-4372-a567-0e02b2c3d479", "radioGroupId": "utg-output-choice-group-unknown_comment_id"} --> Create PR with unit tests - [ ] <!-- {"checkboxId": "6ba7b810-9dad-11d1-80b4-00c04fd430c8", "radioGroupId": "utg-output-choice-group-unknown_comment_id"} --> Commit unit tests in branch `update_` </details> </details> <!-- finishing_touch_checkbox_end --> <!-- tips_start --> --- Thanks for using [CodeRabbit](https://coderabbit.ai?utm_source=oss&utm_medium=github&utm_campaign=DTJ-Tran/pilot_msk_ultrasound_stack&utm_content=1)! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. <details> <summary>❤️ Share</summary> - [X](https://twitter.com/intent/tweet?text=I%20just%20used%20%40coderabbitai%20for%20my%20code%20review%2C%20and%20it%27s%20fantastic%21%20It%27s%20free%20for%20OSS%20and%20offers%20a%20free%20trial%20for%20the%20proprietary%20code.%20Check%20it%20out%3A&url=https%3A//coderabbit.ai) - [Mastodon](https://mastodon.social/share?text=I%20just%20used%20%40coderabbitai%20for%20my%20code%20review%2C%20and%20it%27s%20fantastic%21%20It%27s%20free%20for%20OSS%20and%20offers%20a%20free%20trial%20for%20the%20proprietary%20code.%20Check%20it%20out%3A%20https%3A%2F%2Fcoderabbit.ai) - [Reddit](https://www.reddit.com/submit?title=Great%20tool%20for%20code%20review%20-%20CodeRabbit&text=I%20just%20used%20CodeRabbit%20for%20my%20code%20review%2C%20and%20it%27s%20fantastic%21%20It%27s%20free%20for%20OSS%20and%20offers%20a%20free%20trial%20for%20proprietary%20code.%20Check%20it%20out%3A%20https%3A//coderabbit.ai) - [LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fcoderabbit.ai&mini=true&title=Great%20tool%20for%20code%20review%20-%20CodeRabbit&summary=I%20just%20used%20CodeRabbit%20for%20my%20code%20review%2C%20and%20it%27s%20fantastic%21%20It%27s%20free%20for%20OSS%20and%20offers%20a%20free%20trial%20for%20proprietary%20code) </details> <sub>Comment `@coderabbitai help` to get the list of available commands.</sub> <!-- tips_end -->
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
Clear labels
bug
Something isn't working
documentation
Improvements or additions to documentation
duplicate
This issue or pull request already exists
enhancement
New feature or request
good first issue
Good for newcomers
help wanted
Extra attention is needed
invalid
This doesn't seem right
question
Further information is requested
wontfix
This will not be worked on
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
msk_admin
No Assignees
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: msk_admin/Lumina-MSK#1
No description provided.
Delete Branch "update_"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?