ShifaMind · A Roshan AI product

Concept-grounded ICD-10 coding for clinicians and coders.

ShifaMind reads a clinical note and returns ranked ICD-10 codes with the concept evidence behind each one. Interpretability is enforced architecturally, not bolted on after training.

0.71
Macro-F1
50-code MIMIC-IV
160
Clinical concepts
Grounded vocabulary
113K
Linked admissions
MIMIC-IV labeled corpus
Try it

See ShifaMind code a clinical note.

Pick a scenario, run the coder, and inspect the concepts behind every code. Prerendered for the demo; the platform runs against your own notes.

scenarioAcute on chronic systolic heart failure
Review clinical note · 5 sections
clinical note5 sections

Chief Complaint

Progressive dyspnea on exertion and orthopnea over 1 week.

History of Present Illness

72M with known HFrEF (LVEF 30%) on lisinopril, carvedilol, and spironolactone presents with one week of progressive shortness of breath, bilateral lower-extremity edema, and a 10-lb weight gain. Sleeping in a recliner due to orthopnea. No chest pain or syncope.

Exam

BP 152/88, HR 96, RR 22, SpO2 91% on room air. JVD 12 cm. Bibasilar crackles. 3+ pitting edema in both lower extremities.

Labs & Imaging

BNP 1850 (markedly elevated). Troponin negative. CXR: bilateral pleural effusions and pulmonary vascular congestion.

Assessment & Plan

Acute on chronic HFrEF exacerbation. Started IV furosemide 80 mg, continue home GDMT, daily weights, fluid restriction 1.5 L.

Let's code this note

with ShifaMind, concept-grounded ICD-10 coding.

Benchmark signal

A performance lead built for clinical trust.

ShifaMind pairs the top Macro-F1 in this comparison with concept-mediated evidence that can be inspected alongside every prediction.

MIMIC-IV top-50 · Macro-F1
0.712
Higher is better
+0.063
Lead over GKI-ICD
By design
Evidence on every output
Read the paper
Full scorecard

Six models. One clear leader.

Same evaluation
ShifaMind
0.712
GKI-ICD
0.649
Gemini 2.5 Pro
0.435
View all six models+
Vanilla CBM
0.164
Claude 4.6
0.343
GPT-5.4
0.417
Interpretability

Metrics general LLMs cannot produce.

The concept layer is evaluated alongside the final code, not inferred afterward.

CSTPR
0.704
Concept-supported true positives

Correct diagnoses that also carry relevant activated concept evidence.

CIM
1.314
Concept influence magnitude

How much the concept-grounded representation contributes at the prediction boundary.

CCR
0.836
Concept-conditioned recall

Recall when the relevant clinical concept is present in the case.

What ShifaMind does

Three primitives, designed to be defended.

The demo above shows the first two. Grounded discussion keeps the same evidence in view.

01

Predict

Ranked ICD-10 codes with confidence and considered alternatives.

02

Explain

Activated clinical concepts make the decision inspectable.

03

Discuss

Grounded follow-up inside the note context, without off-topic generation.

Breadth

Coverage across code systems, note types, and specialties.

Code systems
  • ICD-10-CMLive
  • ICD-10-PCSRoadmap
  • CPTRoadmap
  • SNOMED CTRoadmap
Note types
  • Discharge summariesLive
  • Inpatient progress notesLive
  • ED notesLive
  • Outpatient encountersRoadmap
Specialties
  • CardiologyLive
  • PulmonologyLive
  • Emergency medicineLive
  • EndocrinologyLive
  • OncologyRoadmap
  • PsychiatryRoadmap
Security & compliance

Clinical AI you can deploy with confidence.

Enterprise safeguards designed for sensitive clinical workflows.

HIPAA-ready
BAA-eligible deployments
Encrypted
TLS 1.3 · AES-256
Auditable
Concepts and evidence logged
Your data stays yours
No training without opt-in
FAQ

Questions we get a lot.

ShifaMind · ready to evaluate

Bring your notes. See the concepts behind every code.

Try pasted notes today, or talk to us about an evaluation path for your workflow.