# greeshmita-23/sentinelrx architecture

> Architecture diagram of the GitHub repository greeshmita-23/sentinelrx, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-14). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/greeshmita-23/sentinelrx
- Repository: https://github.com/greeshmita-23/sentinelrx

## Overview

SentinelRx is a notebook-based healthcare ML security project. Its documented workflow audits medication-adherence data, detects and removes target leakage, trains and evaluates models, then explores explainability, attacks, and defenses. The README documents these capabilities and a sequential workflow, but the notebook was not excerpted, so internal calls and implementation details cannot be verified; edges below reflect only documented relationships.

Main entry points: [`SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb) (Model evaluation), [`healthcareTrain.csv`](https://github.com/greeshmita-23/sentinelrx/blob/main/healthcareTrain.csv) (Training data), [`healthcareTest.csv`](https://github.com/greeshmita-23/sentinelrx/blob/main/healthcareTest.csv) (Test data), [`secure_adherence_model.joblib`](https://github.com/greeshmita-23/sentinelrx/blob/main/secure_adherence_model.joblib) (Trained model)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Data integrity"]
  node_train_data[("Training data")]
  node_test_data[("Test data<br/>[healthcareTest.csv]")]
  node_audit["Dataset audit"]
  node_leakage["Leakage detection"]
  node_features["Feature engineering"]
end

subgraph group_model["Model development"]
  node_training["Model training"]
  node_benchmark["Model benchmarking"]
  node_calibration["Model calibration"]
  node_evaluation["Model evaluation"]
  node_explainability["Feature importance"]
  node_model_artifact["Trained model"]
end

subgraph group_redteam["Security evaluation"]
  node_adversarial["Adversarial examples"]
  node_membership["Membership inference"]
  node_poisoning_attack["Label-flip poisoning"]
  node_backdoor["Backdoor simulation"]
  node_extraction["Model extraction"]
end

subgraph group_defense["Privacy and defense"]
  node_smoothing["Randomized smoothing"]
  node_dp["Differential privacy"]
  node_sanitization["Poisoning sanitization"]
  node_firewall["Runtime firewall"]
end

node_user(("Analyst"))

node_user -->|"runs notebook"| node_audit
node_train_data -->|"audits"| node_audit
node_test_data -.->|"audits"| node_audit
node_audit -->|"checks leakage"| node_leakage
node_leakage -->|"removes leakage"| node_features
node_features -->|"prepares features"| node_training
node_training -->|"benchmarks classifiers"| node_benchmark
node_benchmark -.->|"calibrates model"| node_calibration
node_calibration -.->|"evaluates model"| node_evaluation
node_training -->|"evaluates model"| node_evaluation
node_training -.->|"saves model"| node_model_artifact
node_evaluation -->|"explains predictions"| node_explainability
node_evaluation -->|"evaluates attack"| node_adversarial
node_evaluation -->|"measures privacy attack"| node_membership
node_evaluation -->|"evaluates attack"| node_poisoning_attack
node_evaluation -->|"evaluates attack"| node_backdoor
node_evaluation -->|"evaluates attack"| node_extraction
node_adversarial -.->|"motivates defense"| node_smoothing
node_membership -.->|"motivates defense"| node_dp
node_poisoning_attack -->|"motivates defense"| node_sanitization
node_training -->|"protects training"| node_dp
node_evaluation -->|"guards runtime inputs"| node_firewall

click node_train_data "https://github.com/greeshmita-23/sentinelrx/blob/main/healthcareTrain.csv"
click node_test_data "https://github.com/greeshmita-23/sentinelrx/blob/main/healthcareTest.csv"
click node_audit "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_leakage "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_features "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_training "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_benchmark "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_calibration "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_evaluation "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_explainability "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_adversarial "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_membership "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_poisoning_attack "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_backdoor "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_extraction "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_smoothing "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_dp "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_sanitization "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_firewall "https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb"
click node_model_artifact "https://github.com/greeshmita-23/sentinelrx/blob/main/secure_adherence_model.joblib"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_train_data,node_test_data,node_audit,node_leakage,node_features toneBlue
class node_training,node_benchmark,node_calibration,node_evaluation,node_explainability,node_model_artifact toneAmber
class node_adversarial,node_membership,node_poisoning_attack,node_backdoor,node_extraction toneMint
class node_smoothing,node_dp,node_sanitization,node_firewall toneRose
class node_user toneIndigo
```

## Components

### Data integrity

- [Training data](https://github.com/greeshmita-23/sentinelrx/blob/main/healthcareTrain.csv): `healthcareTrain.csv`
- [Test data](https://github.com/greeshmita-23/sentinelrx/blob/main/healthcareTest.csv): `healthcareTest.csv`
- [Dataset audit](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Leakage detection](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Feature engineering](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`

### Model development

- [Model training](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Model benchmarking](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Model calibration](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Model evaluation](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Feature importance](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Trained model](https://github.com/greeshmita-23/sentinelrx/blob/main/secure_adherence_model.joblib): `secure_adherence_model.joblib`

### Security evaluation

- [Adversarial examples](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Membership inference](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Label-flip poisoning](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Backdoor simulation](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Model extraction](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`

### Privacy and defense

- [Randomized smoothing](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Differential privacy](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Poisoning sanitization](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`
- [Runtime firewall](https://github.com/greeshmita-23/sentinelrx/blob/main/SentinelRx%20-%20Adversarially-Robust%20%26%20Privacy-Preserving%20Medication-Adherence%20ML.ipynb): `SentinelRx - Adversarially-Robust & Privacy-Preserving Medication-Adherence ML.ipynb`

### Other components

- **Analyst**

## Connections

- Analyst → Dataset audit: runs notebook (evidence: `README.md`)
- Training data → Dataset audit: audits (evidence: `README.md`)
- Test data → Dataset audit: audits
- Dataset audit → Leakage detection: checks leakage (evidence: `README.md`)
- Leakage detection → Feature engineering: removes leakage (evidence: `README.md`)
- Feature engineering → Model training: prepares features (evidence: `README.md`)
- Model training → Model benchmarking: benchmarks classifiers (evidence: `README.md`)
- Model benchmarking → Model calibration: calibrates model
- Model calibration → Model evaluation: evaluates model
- Model training → Model evaluation: evaluates model (evidence: `README.md`)
- Model training → Trained model: saves model
- Model evaluation → Feature importance: explains predictions (evidence: `README.md`)
- Model evaluation → Adversarial examples: evaluates attack (evidence: `README.md`)
- Model evaluation → Membership inference: measures privacy attack (evidence: `README.md`)
- Model evaluation → Label-flip poisoning: evaluates attack (evidence: `README.md`)
- Model evaluation → Backdoor simulation: evaluates attack (evidence: `README.md`)
- Model evaluation → Model extraction: evaluates attack (evidence: `README.md`)
- Adversarial examples → Randomized smoothing: motivates defense
- Membership inference → Differential privacy: motivates defense
- Label-flip poisoning → Poisoning sanitization: motivates defense (evidence: `README.md`)
- Model training → Differential privacy: protects training (evidence: `README.md`)
- Model evaluation → Runtime firewall: guards runtime inputs (evidence: `README.md`)