🔬 IEEE Research Implementation

ML-Based Symptom
Pattern Classification System

An advanced machine learning system for primary health screening that evaluates symptom patterns and proposes potential health conditions using NLP preprocessing and ensemble classification — achieving 91.7% diagnostic accuracy.

91.7%
M1 Accuracy
0.93
NDCG Score
65.3%
Improvement
40+
Diseases
⚙️ Architecture

How It Works

Our 4-stage diagnostic pipeline transforms raw symptom descriptions into ranked differential diagnoses using state-of-the-art NLP and ensemble ML.

1
📝

Text Input

Raw unstructured patient symptom descriptions are fed into the system as clinical vignettes or free-text input.

2
🔤

NLP Preprocessing

Tokenization, stopword removal, lemmatization, and TF-IDF extraction convert text into mathematical feature vectors.

3
🤖

ML Classification

Ensemble model (Random Forest + SVM + Gradient Boosting) cross-references symptoms against known disease patterns.

4
📊

Diagnosis Output

Generates ranked differential diagnoses with confidence scores, severity levels, and recommended next steps.

🩺 Interactive Tool

Symptom Checker

Describe your symptoms in natural language or select common symptoms below. Our ML model will analyze the patterns and suggest potential conditions.

Describe Your Symptoms

🏥 Patient Context (Optional — EHR Integration)
Diabetes Hypertension Heart Disease Asthma Obesity Smoker Immunocompromised Pregnancy
🌡 Fever 🤕 Headache 😷 Cough 💔 Chest Pain 😮‍💨 Breathing Difficulty 😴 Fatigue 🤢 Nausea 😵 Dizziness 🦴 Joint Pain 🗣 Sore Throat 🔴 Rash 🤰 Abdominal Pain ⚖️ Weight Loss 🌙 Night Sweats 🤮 Vomiting 🦵 Swelling
🔬
Enter your symptoms and click analyze
to see diagnostic predictions
📊 Research Data

Comparative Performance Results

Benchmarked against rule-based tools and experienced physicians using 400 medically validated clinical vignettes.

91.7%
M1 Accuracy
Top-1 diagnostic accuracy of our ML system
65.3%
Improvement
Over lowest-performing rule-based tool
0.93
NDCG Score
Surpassing physicians (0.82) in ranking

M1 Accuracy & F1-Score Comparison

Overall Performance Radar

F1-Score Formula (Eq. 1)

F₁ = 2 × (Precision × Recall) / (Precision + Recall)

The F1-score is the harmonic mean of precision and recall, penalizing systems that disproportionately output false positives or false negatives.

🔬 Methodology

Evaluation Parameters

Our testing protocol employed rigorous, standardized methodology to ensure objectivity and clinical relevance.

S.No Parameter Description Remarks
1. Test Dataset 400 clinical vignettes Peer-reviewed
2. Human Benchmark 3 primary care physicians 16.6 years avg. experience
3. System Comparison ML vs existing checkers Identifies gaps
4. Accuracy Metrics M1, F1-score, NDCG Standardized
5. Overall Objective Improve disease triage accuracy Highly effective

Comparative Performance Results (Table II)

Evaluation Entity M1 Accuracy F1-Score NDCG
Lowest Rule-Based Tool 32.4% 0.41 0.45
Human Physicians 88.2% 0.89 0.82
Proposed ML System 91.7% 0.87 0.93
📐 Ranking Quality

NDCG Evaluation Dashboard

Normalized Discounted Cumulative Gain measures how well our system ranks the differential diagnoses — the most critical metric for clinical utility.

Live Model NDCG

NDCG@5 Score

NDCG Comparison

Loading...
0.9–1.0 Excellent
0.7–0.9 Good
0.5–0.7 Fair
<0.5 Poor

NDCG Formula

NDCG = DCG / IDCG   where   DCG = Σ(relᵢ / log₂(i+1))

DCG sums relevance scores discounted by rank position. IDCG is the ideal DCG (perfect ranking). Their ratio gives a normalized score between 0 and 1.

⚖️ Fairness Analysis

Bias Detection Panel

Analyzing model accuracy across disease categories to identify potential data biases — a key focus area from the paper's future work recommendations.

Accuracy by Disease Category

Loading bias data...

Bias Summary

Loading...
👥 Team

Research Team

Symbiosis Institute of Technology, Hyderabad Campus — Symbiosis International University

AM
Adnan Mohammed
Researcher
SIT Hyderabad
NL
Narala Lakshman Reddy
Researcher
SIT Hyderabad
PS
Pulegari Shashi Kiran Reddy
Researcher
SIT Hyderabad
KS
Kiran Siripuri
Faculty Advisor
SIT Hyderabad
SP
Sai Prashanth Mallellu
Corresponding Author
SIT Hyderabad
RA
Rajanikanth Aluvalu
Faculty Advisor
SIT Hyderabad