1 of 6

TITLE PAGE�

  • Team Name - Phoenix

​

1

  • Problem Statement Title- Al-Based Predictive Modelling for Early Forecasting of Bovine Mastitis in lndian Dairy Farms

​

  • Team Member – DIYA BAJPAI

JYOTY KUSHWAHA

ARK SHARMA

​

  • DESCRIPTION - AI-Based Predictive Modelling for Early Forecasting of Bovine Mastitis in lndian Dairy Farms

Phoenix

2 of 6

PROBLEM AND TARGET USERS

CHALLANGES

STRATEGY

POOR INTERNET

CONNECTIVITY

DELAY / LIMITED

DATA

FARMER & VET

ACCESSIBILITY

LANGUAGE & DIGITAL

LITERACY BARRIER

DATA ACCURACY

CONCERNS

HIGH COST

EQUIPMENT

DATA MANAGEMENT

& TRACKING

Offline -first Analysis +

Local Storage +

Auto sync when online

Real-time CMT, EC &

Temperature measurement

+ Instant on-device analysis

Low-cost, portable device

using affordable sensors

& ESP32-CAM

Al/ML model + CMT image

analysis + Multi-parameter

validation

Simple UI, local language

support + Step-by-step

guided process

Mobile App + OLED Display

+ Offline results + Easy to

use in field conditions

Cloud sync + Dashboard + History & trend tracking

GauDhara

Al Based Early Mastitis

Risk Detection System

TECHNICAL

FEASIBILITY

OPERATIONAL

FEASIBILITY

SOCIAL

FEASIBILITY

ECONOMIC

FEASIBILITY

SCALABILITY

DATA

FEASIBILITY

RELIABILITY &

ACCURACY

Easy to operate,

portable & designed

for field usage by

farmers & vets.

Al Based Early Mastitis

Risk Detection System

Modular design &

cloud architecture

enables scale to

thousands of users.

Secure local storage+ Cloud sync ensures

safe & scalable

data handling.

Multi-parameter

analysis + Al model

ensures reliable

results.

Low-cost device

reduces expenses & increases

affordability.

Improves animal health, milk quality

& farmer income.

ROADMAP

Statewide

network

District

Rollout

Mobile App

Deployment

Data

integration

Database

Integration

Backend

Development

Initial Prototype

Phoenix

3 of 6

3

SOLUTION

Phoenix

4 of 6

Early Mastitis Risk� Detection & Monitoring System for Dairy Cattle

4

Phoenix

Detect Early - Treat Timely - Save Lives

WORKING & IMPLEMENTATION FLOW

Milk Sample Collection

From Dairy Cattle

CMT Test + Image Capture

ESP32-CAM captures CMT reaction

Electrical Conductivity (EC)

Measured using SS electrodes

Temperature Measurement

Using DS18B20 sensor

Data Processing

Preprocessing & Feature Extraction

Al / Rule Engine

Risk Prediction Model

Display & Alerts

OLED Display & Flutter App Notification

Risk Classification

NORMAL / SUSPICIOUS / HIGH RISK

1

5 of 6

TECHNICAL APPROACH

Start

Collect Multi-Modal Data

Activity

(MPU6050)

Temperature

(DS18B20)

Milk EC

CMT / Optical

Signal

Historical

Data

Al Risk Analysis

Risk Score ≥

Early-Risk Threshold?

Risk Score ≥

High-Risk Threshold?

Early Risk

Repeat Milk

Screening

Normal

Continue Routine

Monitoring

High Risk

Veterinary Confirmation

Recommended

End

Feasibility Decision Tree

YES

NO

NO

YES

SmartMPU6050 (Activity) Collar

Edge Device

(ESP32)

Data

Transmission

Combines multiple signals:

Multi-Modal

Risk Engine

  • Animal Activity
  • Temperature Trend
  • Milk EC
  • CMT/Optical Signal

Risk Output

(7-14 Days Ahead)

  • No Risk
  • Low Risk
  • Moderate

Risk

  • High Risk

Mobile App

(Farmer / Vet)

Web Dashboard (Admin / Vet)

SYSTEM Architecture

Data Preprocessing & Feature Extraction

Smart Collar (Animal Context)

MPU6050 (Activity)

DS18820 (Temperature)

Optional Optical Sensor

Battery & Local Storage

Smart Milk Testing Box

EC Measurement

Temperature

CMT Optical Signal

Optional Optical/Microscopy

ESP32-based Processing

Sensor Data Acquisition

Preprocessing

Noise Filtering

Feature Extraction

Local Storage

Wireless Sync

(BLE / Wi-Fi)

Send data to

Cloud / Backend

(when available)

Local Storage

(Offline Mode)

  • Historical Data

Application Layer

  • Mobile App (Android)
  • Web Dashboard

(React + TypeScript)

  • Real-time Updates
  • Alerts & Notifications
  • History & Reports
  • Farm & Animal

Management

SMART DATA COLLECTION

On-Cow Layer

Edge Layer

ESP32)

Communication

Cloud / Backend

Smart Collar

Milk Testing Box

Environmental Sensors

  • Data Acquisition
  • Preprocessing
  • Local Storage
  • Anomaly Check

Wi-Fi/ BLE

(when available)

Offline Mode

(Local Storage)

  • API Gateway
  • Database (PostgreSQL)
  • File Storage
  • ML Model Service

(MPU6050, DS18820)

(EC, Temp, CMT/Optical)

(Temp, Humidity, Air Quality)

Technical Workflow

TECH STACK

6 of 6

@SIH Idea submission- Template

6

POTENTIAL AND NEXT STEPS

Phoenix