Journal Article

Field validation of deep learning based Point-of-Care device for early detection of oral malignant and potentially malignant disorders

G., Shubha, Li, Shaobai, O.S., Kathryn, G., Keerthi, Ramesh, Rohan Michael, Kuriakose, Moni A., Vaibhavi, Daksha, Sunny, Sumsum P., R., Vidya Bhushan, Kolur, Trupti, Rajeev, Surya, A.R., Subhashini, Suresh, Amritha, Liang, Rongguang, Mukhia, Nirza, Shetty, Vivek, Leivon, Shirley T., Sigamani, Alben, Smith, Petra Wilder, Patrick, Sanjana, Imchen, Tsusennaro, Pednekar, Sneha, Pillai, Vijay, Mendonca, Pramila, Birur N., Praveen, Song, Bofan, Banik, Ankita Dutta

2022

Overview

Early detection of oral cancer in low-resource settings necessitates a Point-of-Care screening tool that empowers Frontline-Health-Workers (FHW). This study was conducted to validate the accuracy of Convolutional-Neural-Network (CNN) enabled m(mobile)-Health device deployed with FHWs for delineation of suspicious oral lesions (malignant/potentially-malignant disorders). The effectiveness of the device was tested in tertiary-care hospitals and low-resource settings in India. The subjects were screened independently, either by FHWs alone or along with specialists. All the subjects were also remotely evaluated by oral cancer specialist/s. The program screened 5025 subjects (Images: 32,128) with 95% (n = 4728) having telediagnosis. Among the 16% (n = 752) assessed by onsite specialists, 20% (n = 102) underwent biopsy. Simple and complex CNN were integrated into the mobile phone and cloud respectively. The onsite specialist diagnosis showed a high sensitivity (94%), when compared to histology, while telediagnosis showed high accuracy in comparison with onsite specialists (sensitivity: 95%; specificity: 84%). FHWs, however, when compared with telediagnosis, identified suspicious lesions with less sensitivity (60%). Phone integrated, CNN (MobileNet) accurately delineated lesions (n = 1416; sensitivity: 82%) and Cloud-based CNN (VGG19) had higher accuracy (sensitivity: 87%) with tele-diagnosis as reference standard. The results of the study suggest that an automated mHealth-enabled, dual-image system is a useful triaging tool and empowers FHWs for oral cancer screening in low-resource settings.

• Publisher: Nature Publishing Group UK,Nature Publishing Group,Nature Portfolio
• Subject: 631/114/1305 / 631/114/1564 / 631/67/1536 / 631/67/1665 / 631/67/2195 / 631/67/2321 / 631/67/2322 / 692/308/2779 / 692/308/575 / 692/699/3020 / 692/699/67 / Accuracy / Biopsy / Cancer screening / Cell Phone / Deep Learning / Diagnosis / Early Detection of Cancer - methods / Empowerment / Histology / Humanities and Social Sciences / Humans / Lesions / Medical screening / Mouth Neoplasms - diagnosis / Mouth Neoplasms - pathology / multidisciplinary / Neural networks / Oral cancer / Point-of-Care Systems / Science / Science (multidisciplinary) / Telemedicine - methods