Official websites use .gov

A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS

A lock ( 🔒 ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Need Help

SCH: INT: Deep Causal Learning for Diagnosis of Chronic Headache and Facial Pain

Award Number: R01DE036983
ORGANIZATION: NATIONAL INSTITUTE OF DENTAL AND CRANIOFACIAL RESEARCH
OPDIV: NIH
AWARD CLASS: DISCRETIONARY
AWARD ACTIVITY TYPE: SCIENTIFIC/HEALTH RESEARCH (INCLUDES SURVEYS)
PERIOD OF PERFORMANCE START DATE: 09/18/2026
PERIOD OF PERFORMANCE END DATE: 07/31/2029

Group Awards By:

View Award Description

SCH: INT: Deep Causal Learning for Diagnosis of Chronic Headache and Facial Pain - The accurate diagnosis of chronic headache and facial pain is severely hindered by the condition’s immense complexity. Symptoms emerge from a tangled web of direct pathological triggers, systemic comorbidities, and psychological and socioeconomic factors. Consequently, more than 30% of patients experience prolonged diagnostic uncertainty, frequently consulting multiple specialists, only to remain trapped in a frustrating cycle of trial-and-error treatments. To break this diagnostic bottleneck, this project pioneers a foundational Deep Dynamic Causal Learning (DCL) framework leveraging a massive retrospective longitudinal cohort (N > 8,500). This research program will be accomplished through three specific aims: first, to eliminate temporal aggregation bias without combinatorial explosion, we cast structure learning as a continuous optimization problem, imposing continuous-time acyclicity constraints directly over the functional derivatives of Neural Temporal Point Processes (NTPPs) to extract a robust event-driven Granger causal graph. Second, to capture the true underlying disease mechanisms, we construct a deep Temporal Structural Causal Model (SCM), intersecting the learned event-graph as a soft structural prior scaffold with the sparse mechanics of clinical interventions. This continuous-time SCM is optimized using a novel hybrid Bayesian architecture that unifies parameter-efficient normalizing flows with amortized variational inference to propagate uncertainty and map invariant counterfactual equations under violated causal sufficiency. Third, to achieve clinical translation, we embed these models into an interactive software prototype for real-time counterfactual reasoning. This framework will be rigorously validated against a multi-institutional longitudinal cohort (N > 8,500), mathematically framing complex clinical treatments as macro-interventions to prove out-of-distribution generalizability and achieve a measurable 50% reduction in patient misdiagnosis rates.


Issue Date FY	Funding FY	Legal Entity Name	Legal Entity Address	Legal Entity City	Legal Entity State	Legal Entity Zip Code	Legal Entity COUNTY	Legal Entity COUNTRY	Assistance Listing	Award Code	Budget Year	Action Date	Action Type	Action Amount

Issue Date FY: 2026 ( Subtotal = $259,551 )
2026	2026	ROCHESTER INSTITUTE OF TECHNOLOGY	1 LOMB MEMORIAL DR	ROCHESTER	NY	14623	MONROE	USA	Oral Diseases and Disorders Research	000	1	9/17/2026	NEW	$259,551
														Subtotal = $259,551

Grand Total All Awards = $259,551

Top

All Categories

About

Search

Reports

Data Submission

Award Information

SCH: INT: Deep Causal Learning for Diagnosis of Chronic Headache and Facial Pain

Award Number: R01DE036983

ORGANIZATION: NATIONAL INSTITUTE OF DENTAL AND CRANIOFACIAL RESEARCH

OPDIV: NIH

AWARD CLASS: DISCRETIONARY

AWARD ACTIVITY TYPE: SCIENTIFIC/HEALTH RESEARCH (INCLUDES SURVEYS)

PERIOD OF PERFORMANCE START DATE: 09/18/2026

PERIOD OF PERFORMANCE END DATE: 07/31/2029

Federal Websites

Department of Health & Human Services

HHS Operating Divisions

HHS Staff Divisions

Download A Document Viewer