ZeptAI logoResearch-led healthcare AIZEPTAI
Scientific Discovery

Scientific Inquiry &Research at ZeptAI

We investigate artificial intelligence and quantum-computing methods that can address complex real-world problems. Our approach combines mathematical formulation, reproducible experimentation, strong classical baselines, responsible evaluation and pathways toward practical implementation.

Artificial Intelligence and Healthcare

Our healthcare AI research focuses on improving how clinical and patient information is captured, organized, explained and used. The work emphasizes human oversight, reliability, privacy and clear limitations rather than autonomous diagnosis.

  • Conversational healthcare systems
  • Clinical information extraction and summarization
  • Trustworthy and explainable AI
  • Uncertainty estimation and calibrated prediction
  • Efficient AI for low-resource settings
  • Privacy-aware and human-in-the-loop systems
Explore Product Translation

Quantum Computing

ZeptAI is developing research capability in quantum computing with emphasis on hybrid quantum-classical methods. The objective is to identify problem structures where quantum algorithms may contribute practical value, while maintaining transparent comparison with strong classical methods.

Quantum optimization (QUBO/Ising)
Variational algorithms & QAOA
Hybrid solver design
Targeted Quantum ML
Noise-aware mitigation
Resource estimation
Classical baselines
Applied healthcare & energy studies

Credibility Standard: Our research does not assume that quantum methods are always superior. Each study is designed to determine when quantum, classical or hybrid approaches are most appropriate.

Explore Quantum Directions
Active Studies

Ongoing Quantum Research

These tracks represent research initiatives our team has active, ongoing programs around.

Ongoing Research

Hybrid Quantum-Classical Optimization

Designing adaptive workflows that formulate, decompose and solve constrained optimization problems using classical and quantum methods.

Concept and Benchmark Development

Quantum Algorithms for Resilient Systems

Studying quantum and quantum-inspired optimization for scheduling, allocation, network resilience and recovery planning.

Experimental Research

Reliable Quantum Machine Learning

Evaluating small variational quantum models under noise, uncertainty and limited-data conditions against compact classical baselines.

Method Development

Quantum Resource and Readiness Assessment

Developing methods to estimate qubits, circuit depth, noise sensitivity, runtime and practical suitability before selecting a quantum solver.

Publications

Published Research & References

We never place manuscripts under review inside the published category. We do not display journal impact factors as the main evidence of quality.

Full Index
Publisheddoi:10.5565/rev/elcvia.1804

Integrated CNN Model for Multi-disease Classification Through Chest X-ray Images

D. Diwakar, D. Raj

ELCVIA Electronic Letters on Computer Vision and Image Analysis • 2026

Contribution: Presents an integrated convolutional neural network framework targeting simultaneous multi-disease classification from radiographs.

Healthcare AIOpen reference
Publisheddoi:10.7717/peerj-cs.3602

Conversational Framework for Mental Health Diagnosis

D. Diwakar, D. Raj, A. Prasad, G. Ali, M. ElAffendi

PeerJ Computer Science • 2026

Contribution: Links conversational collection with diagnostic classification in a structured workflow.

Healthcare AIOpen reference
Publisheddoi:10.1007/978-981-99-9621-6_6

DistilBERT-based Text Classification for Automated Diagnosis of Mental Health Conditions

D. Diwakar, D. Raj

Microbial Data Intelligence and Computational Techniques for Sustainable Computing • 2024

Contribution: Explores lightweight transformer models for natural language diagnostics in remote mental health monitoring.

Healthcare AIOpen reference
Publisheddoi:10.5815/ijigsp.2022.02.05

Recent Object Detection Techniques: A Survey

D. Diwakar, D. Raj

International Journal of Image, Graphics and Signal Processing (IJIGSP) • 2022

Contribution: Provides a systematic review of object detection architectures and their operational trade-offs.

Machine LearningOpen reference
Publisheddoi:10.1016/j.engappai.2025.112358

Interpretable Chest X-ray Localization with Principal Components

D. Diwakar, D. Raj, K. S. Kumar, G. Ali, A. Prasad

Engineering Applications of Artificial Intelligence • 2025

Contribution: Demonstrates chest X-ray localization with Principal Components for better clinical explainability.

Explainable AIOpen reference
Under Process

Decomposition Strategies for Constrained Quadratic Unconstrained Binary Optimization

DR. Diwakar, Prabhav, Santosh

Contribution: Proposes a novel decomposition approach for mapping QUBO instances onto hybrid classical-quantum solvers.

Optimization
Under Process

Evaluating Small Variational Quantum Models Under Noise and Limited Data

DR. Diwakar, Prabhav, Santosh

Contribution: Establishes benchmarking parameters for variational circuits against compact classical baselines.

Quantum
Under Process

Uncertainty Estimation and Calibration in Deep Clinical Intake Models

DR. Diwakar, Prabhav, Santosh

Contribution: Introduces temperature scaling for calibrating multi-language disease screening systems.

Machine Learning

Collaborate on Research

We are open to joint research, grant proposals, benchmark development, student projects, academic-industry translation and international collaboration in AI, healthcare technology and quantum computing.

Start Collaboration Form