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

Mobile phone-based deep learning algorithm for oral lesion screening in low-resource settings

Award Number: R33CA274717
ORGANIZATION: NATIONAL CANCER INSTITUTE
OPDIV: NIH
AWARD CLASS: DISCRETIONARY
AWARD ACTIVITY TYPE: SCIENTIFIC/HEALTH RESEARCH (INCLUDES SURVEYS)
PERIOD OF PERFORMANCE START DATE: 06/07/2023
PERIOD OF PERFORMANCE END DATE: 05/31/2028

Group Awards By:

View Award Description

Mobile phone-based deep learning algorithm for oral lesion screening in low-resource settings - An estimated 350,000 new cases of oral cancers and 177,000 deaths occur worldwide every year. Lack of reporting in many global settings means that the incidence rates and prevalence are certainly much higher. Two-thirds of oral cancers occur in low-and middle-income countries (LMICs), with 5-year survival rates of only 10-40%. Particularly high incidence occurs in India, with an estimated 200,000 new cases per year. The poor survival rate in LMICs is due to late diagnosis and treatment. To meet the need for timely screening and early detection of oral cancers in LMIC settings, we will develop and validate a low-cost mobile phone-based imaging device powered by computer vision and deep learning classification algorithms. We are a multi-institutional team comprising of optical imaging and machine learning engineers and oral oncologists, at the University of Arizona, Memorial Sloan Kettering Cancer Center (MSKCC) and Tata Memorial Hospital (TMH, Mumbai). In preliminary studies, our team has developed and tested a dual-mode polarized white light imaging (pWLI) and autofluorescence imaging (AFI) mobile device. Non-expert field healthcare workers read images with low sensitivity of 60%. Additionally, a preliminary deep learning classification algorithm, implemented on a cloud-based server, demonstrated improved sensitivity of 79% and specificity of 82%. Our proposal is to address the key remaining hurdle – improving the reading skills of non-expert field healthcare workers – locally in LMIC settings, which do not have internet and cloud connectivity. We will develop and validate the required deep learning image classification algorithm on a mobile phone, to guide field healthcare workers for triage of oral lesions into benign (patients can go home) versus suspicious (patients referred to clinician for follow up care). In the R21 project, we will develop a mobile deep learning-based oral lesion screening and patient triage algorithm and demonstrate feasibility in a cancer care setting (TMH’s main hospital in Mumbai). In the R33 project, we will test and validate the algorithm in a large study in a field setting (TMH’s regional clinic in Varanasi). Successful completion of this project will deliver urgently needed capabilities to field healthcare workers in LMICs, for early detection and triage of oral lesions, improving early oral cancer detection rates, allowing timely referral to specialists, improving treatment outcomes. Relevance to both domestic US and global patient populations: reverse innovation and impact: oral cancer is prevalent in the US population, as well, with about 54,000 new cases per year. Our studies in India will allow rapid and lower-cost access to a significantly larger patient population for large scale image data acquisition that is crucial to support robust development of machine learning algorithms. Machine learning algorithms developed with a very large amount of image data from India will be robust and can be rapidly re-trained with domestic data from MSKCC, validated in the domestic US patient population and implemented in US clinical settings. Our proposed work in India will thus circle back to benefit patients in the USA (reverse innovation and impact).


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 = $293,131 )
2026	2026	SLOAN-KETTERING INSTITUTE FOR CANCER RESEARCH	1275 YORK AVE	NEW YORK	NY	10065	NEW YORK	USA	Cancer Cause and Prevention Research	002	4	7/14/2026	SUPPLEMENT FOR EXPANSION	$66,340
2026	2026	SLOAN-KETTERING INSTITUTE FOR CANCER RESEARCH	1275 YORK AVE	NEW YORK	NY	10065	NEW YORK	USA	Cancer Cause and Prevention Research	001	4	6/23/2026	NON-COMPETING CONTINUATION	$226,791
2026	2025	SLOAN-KETTERING INSTITUTE FOR CANCER RESEARCH	1275 YORK AVE	NEW YORK	NY	10065	NEW YORK	USA	Cancer Cause and Prevention Research	000	3	1/21/2026	SUPPLEMENT FOR EXPANSION	$0
														Subtotal = $293,131

Issue Date FY: 2025 ( Subtotal = $331,721 )
2025	2025	SLOAN-KETTERING INSTITUTE FOR CANCER RESEARCH	1275 YORK AVE	NEW YORK	NY	10065	NEW YORK	USA	Cancer Cause and Prevention Research	001	3	9/17/2025	SUPPLEMENT FOR EXPANSION	$21,816
2025	2025	SLOAN-KETTERING INSTITUTE FOR CANCER RESEARCH	1275 YORK AVE	NEW YORK	NY	10065	NEW YORK	USA	Cancer Cause and Prevention Research	000	3	9/10/2025	EXTENSION WITH OR WITHOUT FUNDS	$309,905
														Subtotal = $331,721

Grand Total All Awards = $624,852

Top

All Categories

About

Search

Reports

Data Submission

Award Information

Mobile phone-based deep learning algorithm for oral lesion screening in low-resource settings

Award Number: R33CA274717

ORGANIZATION: NATIONAL CANCER INSTITUTE

OPDIV: NIH

AWARD CLASS: DISCRETIONARY

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

PERIOD OF PERFORMANCE START DATE: 06/07/2023

PERIOD OF PERFORMANCE END DATE: 05/31/2028

Federal Websites

Department of Health & Human Services

HHS Operating Divisions

HHS Staff Divisions

Download A Document Viewer