TECH-A-THON 4.
0
TECH-A-BIT
PROBLEM STATEMENTS
HEALTH CARE
CORE
● STATEMENT 1 :- Develop an IoT solution for remote patient monitoring that
continuously tracks patients’ health metrics, transmits data securely, and provides
actionable insights to healthcare providers or caregivers. The system must also be
capable of detecting emergencies and triggering an automated emergency response
protocol.
Description:
Develop a smart IoT-based remote patient monitoring solution that continuously tracks key
health metrics such as heart rate, blood pressure, glucose levels, oxygen saturation, and
respiratory rate using wearable or embedded sensors. The system processes real-time data to
detect abnormalities and automatically generates alerts for healthcare providers, patients, and
caregivers.
In case of emergencies (e.g., cardiac arrest, hypoglycemia), the system instantly notifies
doctors, family members, and nearby emergency services with the patient’s GPS location and
critical health data. It may also activate smart home devices to assist emergency responders.
The platform includes a secure, user-friendly mobile app or dashboard where patients can
monitor their health trends, receive personalized insights, medication reminders, and health tips.
AI and machine learning models provide predictive analysis for early warning of health
complications. A virtual AI assistant can offer guidance, emotional support, and real-time
interaction based on the patient’s condition.
All data is transmitted securely using encryption protocols to ensure privacy and compliance
with healthcare standards.
● STATEMENT 2 :- Develop a comprehensive mental health monitoring solution that
integrates hardware and software to track and improve mental well-being
Description:
The solution could focus on one or more of the following:
# Wearable Devices: Create a wearable device (e.g., smartwatch, wristband) that monitors
physiological indicators associated with mental health (e.g., heart rate variability, sleep patterns,
physical activity) and alerts users to significant changes.
# Mobile Application: Develop an app that allows users to log their mood, stress levels, and
other mental health indicators daily, offering insights and personalized suggestions for coping
strategies.
# Real-Time Monitoring and Alerts: Implement a system that analyzes data from wearables and
mobile apps in real time, providing alerts to users and caregivers if indicators suggest a
potential mental health crisis.
# Virtual Support and Resources: Create a platform that connects users with mental health
resources, including teletherapy options, community support groups, and educational materials
tailored to their needs.
# AI-Powered Insights and Recommendations: Utilize machine learning algorithms to analyze
user data, providing actionable insights and personalized recommendations for improving
mental well-being.
● STATEMENT 3 :- Design a decentralized, AI-powered bio-waste monitoring system in
hospitals to detect infection outbreaks early and predict hospital-acquired infection (HAI)
risk in real time.
Description:
Develop a smart, AI-based system that monitors hospital bio-waste in real time to detect
infection outbreaks early. Using biosensors and edge AI in waste bins, the system analyzes
biological residues to identify rising infection risks (e.g., bacteria, blood traces), predicts
hospital-acquired infections (HAIs), and alerts staff for rapid response. It improves infection
control, ensures waste compliance, and helps prevent secondary outbreaks in healthcare
facilities.
● STATEMENT 4 :- Smart Helmet for Road Hazard Detection and Rider Safety
Description:
The solution could focus on one or more of the following:
# Embedded Camera & Sensors: Captures road visuals and uses an IMU (Inertial Measurement
Unit) to detect vibrations or potholes.
# Fatigue Detection: Tracks eye blinks, head tilt, and movement patterns to identify drowsiness.
# Crash Detection & Alerts: On sensing an impact, it automatically alerts emergency contacts
and shares GPS location.
# Smart Speed Guidance: Uses AI to recommend speed limits based on road conditions and
rider behavior.
# P2P Alerts: Sends real-time road hazard alerts to other nearby smart helmet users.
● STATEMENT 5 :- Modular Bio-Patch for Multi-Organ Stress Detection
Description:
Design a next-gen wearable bio-patch that is modular and flexible, capable of detecting stress
on different internal organs (e.g., heart, kidney, liver) using a combination of bioimpedance,
GSR, skin temperature, and sweat composition analysis. The patch must use low-power
electronics and edge-AI inference to provide real-time risk analysis without relying on cloud
computing.
The device should include swappable or attachable modules based on specific use cases (e.g.,
kidney health in diabetics vs cardiac stress in athletes). The patch must communicate wirelessly
with a smartphone or medical interface while preserving patient privacy. Visual feedback (LED)
and haptic feedback (vibration) for critical alerts are required.
● STATEMENT 6 :- Point-of-Care Diagnostic Lab-on-Chip with Smartphone Integration
Description:
Develop a pocket-sized, hardware-integrated diagnostic device that performs point-of-care
blood/urine tests using microfluidics and colorimetric analysis, all while interfacing with a
smartphone. It should be capable of detecting key health indicators like glucose, creatinine, and
protein levels through simple color change or conductivity tests on a disposable cartridge.
The user inserts a small biological sample (e.g., a finger prick of blood or a urine drop), and the
chip routes it to different chambers containing reagents. An onboard optical sensor (or camera
interface) detects changes and classifies results, which are then transmitted to the user’s
smartphone for visualization, medical storage, and optional remote doctor sharing.
TECH
● STATEMENT 7 :- Counterfeit Drug Detection Using RFID and Blockchain Verification
Description:
Build a software prototype that simulates an RFID-based medicine verification system. When a
medicine's RFID tag is scanned (emulated via QR code or RFID reader), the system checks the
tag ID against a secure backend (e.g., a simulated blockchain or encrypted database) to confirm
authenticity, expiration, and manufacturer source.
You may also implement a web portal for pharmacies/hospitals to verify tags and log
transactions.
● STATEMENT 8 :- Smart Fall Detection Using Motion Sensor Data and Machine
Learning
Description:
Design a software solution that utilizes accelerometer and gyroscope data from wearable
devices or smartphones to detect accidental falls in elderly individuals. The system should
differentiate between normal activities (e.g., sitting, walking) and actual falls using AI/ML models
and trigger alerts to caregivers or emergency services.
The goal is to minimize false alarms while maintaining real-time response. You can use
pre-recorded sensor datasets and simulate incoming data for training and testing the system.
● STATEMENT 9 :- Real-Time Blood Shortage Alert and Donor Mobilization System
Description:
Design and implement a software prototype for an emergency blood donation web
platform. The system should allow hospitals and blood banks to raise real-time alerts
when a specific blood type is in critical shortage. When a request is triggered, the platform
identifies eligible past donors within a specific radius (based on blood group, geolocation, and
donation history) and notifies them via SMS, email, or web push notifications.A hospital-facing
dashboard must show current emergency requests, track donor responses, monitor blood
inventory levels, and allow coordination between hospitals for unit sharing. Security, donor
eligibility logic, and role-based access control are essential components.
You may also simulate geolocation matching using MongoDB geospatial queries or
coordinate-based logic.
● STATEMENT 10 :- Family-Centered Mental Health Monitoring and Support System
Description:
Build a software prototype that simulates a family mental health support platform where
family members can collaboratively monitor and support each other's mental wellness through
shared behavioral insights, mood tracking, and communication pattern analysis. When the
system detects changes in family member behavior, communication patterns, or mood
indicators, it provides family-appropriate intervention strategies and connects them with
professional resources. The system should handle multi-user family accounts with appropriate
privacy controls, age-appropriate interfaces for different family members, and family
communication facilitation tools. Implement a web portal for families to track collective wellness,
share supportive messages, and access mental health resources together.
● STATEMENT 11 :- Early Sepsis Prediction System for ICU Using Real-Time
Multi-Sensor Data Streams
Description:
Develop a software prototype that simulates a real-time ICU monitoring system capable of early
sepsis prediction using multi-sensor data streams such as heart rate, respiratory rate, oxygen
saturation, body temperature, and blood pressure.
The system continuously ingests streaming data (e.g., from simulated sensor feeds or
pre-recorded ICU datasets) and applies an LSTM-based deep learning model to identify early
sepsis indicators based on temporal patterns.
It must provide:
Dynamic risk scores for patients.
Real-time alerts to ICU staff with confidence levels.
Model explanation using SHAP or Grad-CAM for medical transparency.
● STATEMENT 12 :- Post-Stroke Cognitive Rehab Companion Using EEG Attention
Feedback and Adaptive Exercises
Description:
Design a cognitive rehabilitation platform for post-stroke patients using low-cost EEG
headbands (like Muse/Neurosky) to measure real-time attention, mental workload, and fatigue.
The system adjusts rehabilitation exercises (e.g., memory games, visual tracking, audio
prompts) based on the user’s live EEG attention scores and reaction times.
Key Components:
# EEG signal processing pipeline (artifact removal, FFT, alpha/beta band detection).
# Adaptive game logic that increases/decreases difficulty in real-time.
# Patient monitoring dashboard with fatigue and engagement tracking.
# Data sync with remote therapists for weekly progress reviews.
DEFENCE
CORE
● STATEMENT 13 :- Real-Time Gunshot Detection and Localization Device for Soldiers
Description:
Create a compact, wearable gunshot detection system that uses acoustic triangulation to
instantly detect and localize nearby gunfire. The device should analyze sound patterns using a
microphone array and classify the type of firearm, distance, and direction of the shot. On
detection, it must provide silent vibration alerts and visual indicators (LEDs) to notify the soldier
and optionally transmit real-time data to the command center.
This device must be rugged, operate in extreme temperatures, and consume low power while
still performing real-time detection and classification using embedded DSP or lightweight ML
models.
● STATEMENT 14 :- Satellite Blackout Early Warning Sensor for Border Camps
Description:
Design a hardware system that continuously monitors GNSS signal integrity (GPS, IRNSS,
GLONASS) and detects spoofing, jamming, or signal blackout, which are indicators of potential
electronic warfare attacks near border installations. The system should raise alerts to command
centers, store forensic logs, and optionally activate inertial tracking fallbacks for critical units.
The sensor node should track parameters like SNR drops, signal delay anomalies, and false
location shifts. It should also visualize signal health locally for soldiers or communication officers
via an LCD/LED interface.
● STATEMENT 15 :- Design smart glasses with a built-in infrared sensor + laser diode that
can detect thin tripwires (used to trigger hidden IEDs/mines) using reflected light and IR
signature anomalies, then overlay real-time alerts in a simple HUD (heads-up display).
Description:
Design smart tactical glasses with an integrated infrared laser emitter and receiver that can
detect almost invisible tripwires used to trigger landmines or booby traps. These wires, often
made of transparent fishing line or thin copper, reflect IR differently from their surroundings. The
system processes reflections in real time, highlights tripwire presence via a heads-up LED
indicator, and alerts the soldier discreetly through bone conduction or vibration.
● STATEMENT 16 :- Build a network of ultra-low-power, solar-powered sensor nodes that
can be air-dropped or deployed in remote regions (forest, desert, mountain) to detect
movement, vibrations, and radio signals from intruders without GPS or RF tracking.
Description:
Create matchbox-sized sensor nodes with vibration, thermal, and RF-detection capabilities that
form a stealthy sensor mesh when deployed manually or airdropped across forested or border
regions. Powered by small solar panels, each node detects human or vehicle movement, logs
activity, and relays it to the base via low-power LoRa mesh networks. The nodes go into sleep
mode to conserve power and awaken only when triggered.
● STATEMENT 17 :- Develop a digital soldier tag that includes encrypted identity, live
vitals (heart rate, temp), and last known location. If the soldier is unconscious or missing,
the tag can broadcast its final state over secure frequencies.
Description:
Develop a rugged, wearable digital ID tag that monitors basic vitals (heart rate, skin temp) and
tracks the soldier's last known coordinates. If the tag is forcibly removed, it auto-locks or wipes
data. Upon abnormal vitals or injury, it triggers a distress beacon, which can operate even in
disconnected environments using store-and-forward logic or short-range encrypted pulses.
● STATEMENT 18 :- Design a glove-based system with pressure sensors that allows
soldiers to tap out Morse code silently, which is encrypted and transmitted to nearby
teammates via mesh communication in GPS-denied or voice-restricted zones.
Description:
Design a smart glove equipped with pressure sensors on fingers and palm, enabling soldiers to
send Morse code or tap-based commands silently. The glove converts tap patterns into
encrypted text signals and transmits them via LoRa mesh, BLE, or IR pulses to nearby
teammates or field command. Haptic feedback confirms successful transmission.
TECH
● STATEMENT 19 :- Secure Inter-Agency Communication Network for Crisis
Management
Description:
Build a software prototype that simulates a dynamic key distribution system for emergency
response scenarios. When multiple agencies need to establish secure communication channels
during crisis situations, the system automatically creates secure communication groups,
distributes encryption keys, and manages access controls across different clearance levels.
The system should handle dynamic group formation where responders can join/leave secure
channels based on their role and clearance level. Implement a web portal for emergency
coordinators to manage communication groups, monitor active channels, and revoke access
when personnel are reassigned or compromised.
You may also implement features for cross-agency authentication, emergency key escrow for
authorized personnel, and backup communication protocols when primary infrastructure fails.
● STATEMENT 20 :- Encrypted File Sharing Platform with Role-Based Access Control
Description:
Build a software prototype that simulates a secure file sharing system for military personnel
to upload, encrypt, and share classified documents with appropriate access controls and
comprehensive audit trails. When a document is uploaded, the system encrypts it client-side
before storage, assigns access permissions based on user roles (Officer, Sergeant, Private),
and generates secure, time-limited sharing links.
The system should handle role-based authentication, file encryption/decryption, permission
inheritance for folder structures, and real-time audit logging of all file operations. Implement a
web portal for users to manage their files, set granular permissions, and administrators to
monitor access patterns and generate compliance reports.
You may also implement features for secure file deletion, key rotation for stored files, offline
file access capabilities, and emergency access procedures for authorized personnel during
critical operations.
● STATEMENT 21 :- Develop a secure web platform that monitors soldier health, behavior,
and mission performance to predict injury, fatigue, or psychological risk using ML models
and real-time inputs.
Description:
The solution could focus on one or more of the following:
# Health and Performance Data Logging: Allow daily uploads of health checkup data, mission
logs, and stress reports from field units.
# Risk Score Prediction: Use classification models (KNN, Random Forest, SVM) to assign risk
levels (Low/Medium/High) for each soldier.
# Personalized Alerts: Alert commanding officers if a soldier’s health or stress indicators cross a
danger threshold.
# Mood Tracker Interface: Provide soldiers with a mobile-accessible app to report stress, sleep,
and general wellness.
# Data Privacy and Encryption: Secure all records with encryption at rest and during transit ,with
multi-level access control.
# Audit Logs: Log every data access or change with timestamps to ensure traceability and
integrity of sensitive health data.
# Visual Dashboard: Provide commanders with overview heatmaps of unit wellness across
locations and missions.
● STATEMENT 22 :- Design a web-based system that uses natural language processing
to classify intercepted communications or reports as benign, suspicious, or critical, and
visually highlights threats for defense analysts.
Description:
The solution could focus on one or more of the following:
# NLP-Based Threat Classification: Build a machine learning model to classify documents,
messages, or transcripts into risk categories using supervised learning and keyword/entity
detection.
# Named Entity Recognition: Highlight critical terms like locations, people, or weapons using
NER models to assist human analysts.
# Interactive Dashboard: Develop a secure web platform where analysts can upload text inputs,
see visual alerts, and track previous intelligence logs.
# Role-Based Access Control: Allow different types of users (analyst, commander) with limited
permissions for viewing or uploading classified information.
# Model Feedback Loop: Analysts should be able to mark false positives or reclassify threats to
continually improve model performance.
# Multilingual Handling: The system should support Hindi, English, and code-mixed inputs like
Hinglish using multilingual embeddings or transformers.
# Data Storage and Search: Allow fast lookup and history tracking of similar threats from
previous data using indexing or vector similarity
● STATEMENT 23 :- Mission Route Replanning System Using Reinforcement Learning
under Dynamic Threat Conditions
Description:
Design a software prototype that simulates an intelligent mission planning system for unmanned
aerial or ground vehicles operating in hostile or unpredictable environments. The system
receives a real-time tactical map with dynamic threat zones such as surface-to-air missile (SAM)
regions, electronic jamming areas, and terrain obstacles. Using reinforcement learning (e.g.,
DQN or PPO), the agent autonomously navigates from start to destination while adapting to
changes like emerging threats or shifting mission objectives (e.g., rerouting to extract an injured
soldier). The model is trained to maximize mission success while minimizing exposure risk and
resource usage (e.g., fuel). The platform includes a visual dashboard to compare the original vs.
optimized path, simulate real-time map updates, and demonstrate replanning logic based on
mission priority. Optional modules may include fuel constraints, multi-agent coordination, or
sensor failure simulation for testing agent robustness.
● STATEMENT 24 :- Insider Threat Detection in Military Networks using Graph Neural
Networks (GNNs)
Description:
Develop a software system that detects anomalous or high-risk user behavior in secure defense
networks by analyzing system activity logs. The input consists of message exchanges, file
access records, and login metadata. Construct an interaction graph where nodes represent
users and edges encode communication or access patterns, weighted by frequency or
sensitivity. Apply a Graph Neural Network (GNN) architecture such as GraphSAGE or GCN to
classify users as normal or potentially malicious based on their structural position in the graph,
deviations from expected behavior (e.g., based on role), and temporal anomalies. Integrate
temporal data structures (e.g., segment trees or time-indexed logs) to enable efficient real-time
querying of user activity over time windows. A visual dashboard can highlight suspicious nodes,
summarize evidence trails, and provide alerts. Optional extensions include simulating data
poisoning, sleeper agents, or role-hopping behavior to evaluate model robustness under
adversarial conditions.
