Onco-Shikshak

A retrieval-augmented AI platform for adaptive oncology education grounded in clinical practice guidelines

Try Live Demo Read the Paper
4 Clinical Workflows
6 Specialist AI Agents
9 Pedagogy Algorithms
18 Clinical Cases
Product

Clinical Workflows in Action

Four authentic clinical workflows powered by 6 specialist AI agents, FSRS v4 spaced repetition, IRT-adaptive difficulty, and NCCN-grounded RAG.

Onco-Shikshak learning hub dashboard

Learning Hub

Personalized dashboard with active sessions, due flashcards, weak area suggestions, and quick access to all clinical workflows.

Virtual tumor board with 6 specialist AI agents

Virtual Tumor Board

Six specialist AI agents (pathology, radiology, surgical oncology, medical oncology, radiation oncology, supportive care) deliberate sequentially on real clinical cases. Each agent queries its own guideline sources via RAG.

Morning report Socratic clinical reasoning

Morning Report

Socratic dialogue with Dr. Chandra. 6-phase clinical reasoning scaffolding with expertise-adaptive difficulty.

FSRS spaced repetition flashcard review

Spaced Repetition

FSRS v4 scheduling with confidence calibration, interleaved practice across cancer domains, and contextual card variants.

Abstract
Medical oncology education demands continuous mastery of rapidly evolving clinical guidelines—the NCCN alone maintains over 76 guideline documents updated multiple times annually. We describe Onco-Shikshak, an AI-native adaptive learning ecosystem integrating three complementary modules: (1) a dynamic textbook generator producing expertise-level-adapted content grounded in retrieved guideline and textbook evidence, (2) a Socratic virtual preceptor guiding clinical reasoning through progressive disclosure and intentional cognitive friction, and (3) a spaced repetition engine generating atomic flashcards from authoritative sources scheduled via the SM-2 algorithm. The system employs retrieval-augmented generation (RAG) over seven clinical guideline corpora—NCCN, ESMO, ASTRO, ACR, CAP, ClinVar/CIViC, and SSO—with automated citation extraction from grounding metadata. The architectural design is grounded in six established learning science frameworks.
Clinical Workflows

Four Authentic Clinical Learning Environments

Each workflow mirrors a real clinical setting, powered by 6 specialist AI agents and 9 pedagogy algorithms sharing a common NCCN-grounded RAG layer.

🏥

Morning Report

Present cases to Dr. Chandra via Socratic dialogue. 6-phase clinical reasoning scaffolding (problem representation through management justification) with expertise-adaptive difficulty from IRT ability estimation.

🧑‍⚕️

Tumor Board

Six specialist AI agents (pathology, radiology, surgery, medical oncology, radiation oncology, supportive care) deliberate sequentially. Cognitive conflict forces the learner to synthesize disagreeing experts.

🏥

Clinic Day

Manage 3–5 simultaneous patients with random clinical interruptions. SOAP note documentation under time pressure with NCCN-grounded attending feedback.

📚

Spaced Repetition 2.0

FSRS v4 replaces SM-2 for optimal scheduling. Interleaved practice across cancer domains, contextual card variants, confidence calibration, and closed-loop error-to-card generation.

📖

Dynamic AI Textbook

Generates personalized oncology chapters adapted to expertise level. RAG-grounded with citations, related cases linked at chapter end, streamed via SSE.

🎯

Virtual Preceptor

One-on-one with any specialist agent. Scaffolding ranges from direct instruction (struggling) to challenge mode (advanced learners adding clinical noise).

Pedagogical Foundations

Theory-to-Design Mapping

Every architectural decision maps to an established learning science framework.

Theory Principle Design Decision
Anderson (2000) Spacing effect & retrieval practice SM-2 algorithm with guideline cards
Yeo & Fazio (2019) Retrieval for facts; worked examples for procedures Flashcards + case simulations
Lee & Anderson (2013) Expertise reversal effect Three adaptive complexity levels
Tankelevitch (2025) Desirable friction; cognitive protection Socratic questioning in preceptor
Matuschak & Nielsen Learning in authentic context Clinical case immersion
LearnLM (2025) AI-augmented textbook RAG-grounded dynamic generation
Knowledge Grounding

7 Guideline Sources + 2 Textbook Corpora

All generated content is grounded in authoritative clinical sources via Gemini File Search RAG, with automated citation extraction from grounding metadata.

NCCN Guidelines ESMO Guidelines ASTRO Guidelines ACR Appropriateness Criteria CAP Cancer Protocols ClinVar / CIViC SSO Surgical Oncology DeVita 11th Ed. Abeloff 6th Ed.
Preprint

Read the Paper

Full preprint describing the system design, architecture, pedagogical rationale, and technical validation.

Download PDF
Cite

Citation

@article{ashish2026oncoshikshak,
  title={Onco-Shikshak: A Retrieval-Augmented AI Platform
         for Adaptive Oncology Education Grounded in
         Clinical Practice Guidelines},
  author={Ashish},
  journal={arXiv preprint},
  year={2026},
  url={https://inventcures.github.io/onco-shikshak/}
}