SBIR-STTR Award

Direct to Phase II

Two-thirds of oral and oropharyngeal squamous cell carcinoma (OSCC) occur in low- and middle-incomecountries (LMICs), the 5-year survival rate is only 10-40. Enough is already known about the disease and itsprevention for action to be taken. The poor survival rate in LMICs is mainly due to late diagnosis. Therefore, it isimperative to detect precursor and malignant lesions early and expeditiously.To meet the need for technologies that enable oral cancer screening and diagnosis in low resource settings(LRS), Light Research Inc (LRI) will develop, validate, and commercialize a low-cost smartphone-based imagingsystem that provides remote specialist access and triage decision-making guidance tailored to non-specialistuse in LRS.To achieve the project goal, LRI will license technologies from University of Arizona (UA), and partner with UAand Mazumdar-Shaw Cancer Center (MSCC) (Bangalore, India) to develop and validate a multimodal mobileoral imaging system for oral cancer detection and diagnosis in LRS. In the past few years, the project team hasdeveloped and evaluated a dual-mode (polarized white light imaging [pWLI]) and autofluorescence imaging[AFI])) mobile imaging device that specifically addresses critical barriers in LRS to improve oral cancer screening.To address one of the key hurdles in adopting optical imaging techniques for oral cancer screening in LRS, theteam has also developed and evaluated cloud-based and mobile-based deep learning image classificationmethods for guiding patient triage.Since the key techniques proposed in this mobile imaging system have been successfully evaluated for oralcancer screening with 3,000 high-risk population in LRS, the potential of successfully transitioning it to the low-resource regions for accurate, objective and location-resolved detection of oral cancer is very high. The projectobjective will be achieved through three Aims: (1) to optimize a mobile intraoral imaging system for LRS, (2) tooptimize deep learning-based dual-modality image classification methods, and (3) to validate the clinicalusefulness of the mobile oral imaging system for oral cancer screening and triage in LSR.Successful completion of this project will lead to a mobile oral imaging system with deep learning imageclassification method that delivers urgently-needed capabilities to the end users in LRS. LRI will partner withJana Care Inc (New Delhi, India) for low-cost production, with Ergo Healthcare (Mumbai, India) for productdistribution in south and southeast Asian, and with DentalEZ Group for product distribution in America andEurope. The use of this mobile-based screening approach for early detection and triage of oral cancers willeventually improve oral cancer detection rates, treatment outcomes, and quality of life of patients in LRS.

Public Health Relevance Statement:
Project Narrative Light Research Inc (LRI) will develop and commercialize a mobile oral imaging system to address the urgent need of a low-cost, portable, easy to use, and reliable imaging device and deep learning image classification method for oral cancer screening in low resource settings (LRS). It is expected that this mobile based dual- modality screening approach for early detection and triage of oral cancers will eventually improve detection rates, treatment outcomes, and quality of life of patients.

Project Terms:
Americas ; Arizona ; Biopsy ; Malignant Neoplasms ; Cancers ; Malignant Tumor ; malignancy ; neoplasm/cancer ; Classification ; Systematics ; Decision Making ; Diagnosis ; Disease ; Disorder ; Electronics ; electronic device ; Europe ; Goals ; Health ; India ; Learning ; Light ; Photoradiation ; Lighting ; Illumination ; Methods ; Morbidity - disease rate ; Morbidity ; mortality ; Patients ; Production ; Quality of life ; QOL ; Research ; Resources ; Research Resources ; Survival Rate ; Technology ; Telephone ; Phone ; Time ; Triage ; Universities ; Imaging Techniques ; Imaging Procedures ; Imaging Technics ; Treatment outcome ; Healthcare ; health care ; Schedule ; Specialist ; oral lesion ; mouth lesion ; Data Set ; Dataset ; Caring ; malignant mouth neoplasm ; Malignant Oral Cavity Neoplasm ; Malignant Oral Cavity Tumor ; Malignant Oral Neoplasm ; Mouth Cancer ; Oral Cancer ; malignant mouth tumor ; oral cavity cancer ; base ; rural area ; rural location ; rural region ; improved ; Solid ; Clinical ; Malignant - descriptor ; Malignant ; Phase ; Lesion ; Oral cavity ; Buccal Cavity ; Buccal Cavity Head and Neck ; Cavitas Oris ; Mouth ; tongue root ; tongue base ; southeast Asian ; Licensing ; Disease Progression ; Oropharyngeal Squamous Cell Carcinoma ; Oropharyngeal Epidermoid Carcinoma ; oropharynx epidermoid carcinoma ; oropharynx squamous cell carcinoma ; Screening for Oral Cancer ; oral cancer early detection ; cancer prevention ; Internet ; WWW ; web ; world wide web ; dyscrasia ; Dysplasia ; Diagnostic ; mechanical ; Mechanics ; Adopted ; Oral ; Techniques ; System ; Location ; early detection ; Early Diagnosis ; impression ; Performance ; Histopathology ; Participant ; Prevention ; Modality ; Devices ; social ; portability ; cancer diagnosis ; Cell Phone ; Cellular Telephone ; iPhone ; smart phone ; smartphone ; Cellular Phone ; image-based method ; imaging method ; imaging modality ; Address ; Data ; Detection ; Imaging Instrument ; Imaging Tool ; Imaging Device ; Cancer Center ; Cancer Detection ; Cancer Patient ; South Asian ; Process ; Image ; imaging ; intraoral probe ; oral care ; cost ; optical imaging ; optic imaging ; imaging probe ; design ; designing ; innovation ; innovate ; innovative ; handheld mobile device ; mobile device ; prototype ; multimodality ; multi-modality ; community setting ; standard of care ; clinical practice ; flexibility ; flexible ; screening ; mobile application ; mobile app ; mobile device application ; cloud based ; Patient Triage ; imaging system ; low and middle-income countries ; LMIC ; high risk population ; high risk group ; high resolution imaging ; recruit ; deep learning ; Infrastructure ; neural network architecture ; neural net architecture ; deep learning algorithm ; classification algorithm ; remote diagnosis ;