DSM-5 Aligned · NLP + ML · Multi-Stream AI

AI-Powered
Psychiatric Screening

CogniDetectAI fuses structured questionnaires with real-time speech analysis to screen for ADHD, Depression, Anxiety, Autism, and SPCD — in five languages.

Begin ScreeningLearn More
This tool is a screening aid only. Not a substitute for professional diagnosis.
5
Conditions Screened
27
DSM-5 Questions
4
AI Guardrail Layers
5
Languages
What We Screen For

Five Conditions, One Platform

CogniDetectAI uses DSM-5 aligned criteria to screen for the five most common neurodevelopmental and mood disorders.

ADHD

DSM-5

Attention Deficit Hyperactivity Disorder

Persistent patterns of inattention, hyperactivity, and impulsivity that interfere with daily functioning.

Depression

DSM-5

Major Depressive Disorder

Persistent low mood, loss of interest, and reduced energy that significantly impacts quality of life.

Anxiety

DSM-5

Generalised Anxiety Disorder

Excessive, uncontrollable worry about multiple life domains accompanied by physical symptoms.

Autism

DSM-5

Autism Spectrum Disorder (ASD)

Challenges in social communication and interaction, alongside restricted or repetitive behaviours.

SPCD

DSM-5

Social Pragmatic Communication Disorder

Difficulty using verbal and non-verbal communication for social purposes without ASD-specific behaviours.

Differential

DSM-5

ASD · SPCD Differential Diagnosis

The AI automatically distinguishes between ASD and SPCD using the DSM-5 logic tree — a clinically significant differentiation.

The Process

How CogniDetectAI Works

A dual-stream AI pipeline combining structured questionnaires with natural language understanding for higher accuracy.

01
Stream A — Structured

Answer the Questionnaire

27 DSM-5 aligned questions scored on a 5-point frequency scale. Model evaluates your responses and flags potential risk domains.

ML on 27 DSM-5 items
02
Stream B — Unstructured

Voice & Text Interview

AI-Based follow-up questions based on your symptom flags. Answer by voice or text.

NLP analysis
03
Meta-Fusion Ensemble

Fusion Report

A trained model combines both streams to produce a final diagnosis and personalised suggestions.

Meta-Fusion with blend
Peer-Reviewed Research

Accepted and TBP at IEEE INSECT-2026

CogniDetectAI was developed as a BE final-year project and accepted for publication at the IEEE International Conference on Intelligent and Sustainable Electronics and Computing Technologies, May 2026.

Conference Paper2026AI · Mental Health

CogniDetectAI: An AI-Powered Clinical Decision Support System for Psychiatric Screening Using Dual-Stream Multimodal Fusion

Sankalp Indish, Dr. Monika Dangore, Aishwarya Borse, Rashi Madne

IEEE INSECT-2026 — Intelligent & Sustainable Electronics and Computing Technologies · May 2026

Authors

S
Sankalp IndishLead Developer
Team Lead & Principal Developer
Complete end-to-end AI in SDLC — deployment, model maintenance, and full research paper authorship with self-made diagrams. Architecting the system with continuous maintenance and updates.
D
Dr. Monika Dangore
Research Supervisor & Project Mentor
Academic supervision, clinical methodology review, pilot dataset development for RF-model with evaluation, drafting the DSM-5 Aligned Questionnaire and IEEE INSECT-2026 submission guidance.
A
Aishwarya Borse
Research Contributor
Project diagrams, report and presentation edits, and documentation.
R
Rashi Madne
Research Contributor
Version 1 project paper and literature survey.
5
Disorder Classes
27
Clinical Questions
~600MB
Model Payload
5 lang
Multilingual
Ready to begin?

Take the Free Screening

Complete anonymously in 10–15 minutes. Supports English, Hindi, Marathi, German, and Mandarin Chinese.

Begin Screening→