TY - JOUR
T1 - Multi-task network for automated analysis of high-resolution endomicroscopy images to detect cervical precancer and cancer
AU - Brenes, David
AU - Barberan, C. J.
AU - Hunt, Brady
AU - Parra, Sonia G.
AU - Salcedo, Mila P.
AU - Possati-Resende, Júlio C.
AU - Cremer, Miriam L.
AU - Castle, Philip E.
AU - Fregnani, José H.T.G.
AU - Maza, Mauricio
AU - Schmeler, Kathleen M.
AU - Baraniuk, Richard
AU - Richards-Kortum, Rebecca
N1 - Funding Information:
This work was supported by National Cancer Institute : R01 CA251911 , R01 CA186132 , UH2/UH3 CA189910 , P30 CA016672 ; NSF grants: CCF-1911094 , IIS-1838177 , and IIS-1730574 ; ONR grants N00014–18-12571 , N00014–20-1–2534 , N00014–18-1–2047 , and MURI N00014–20-1–2787 ; AFOSR grant FA9550–18-1–0478 ; and a Vannevar Bush Faculty Fellowship .
Publisher Copyright:
© 2022 Elsevier Ltd
PY - 2022/4
Y1 - 2022/4
N2 - Cervical cancer is a public health emergency in low- and middle-income countries where resource limitations hamper standard-of-care prevention strategies. The high-resolution endomicroscope (HRME) is a low-cost, point-of-care device with which care providers can image the nuclear morphology of cervical lesions. Here, we propose a deep learning framework to diagnose cervical intraepithelial neoplasia grade 2 or more severe from HRME images. The proposed multi-task convolutional neural network uses nuclear segmentation to learn a diagnostically relevant representation. Nuclear segmentation was trained via proxy labels to circumvent the need for expensive, manually annotated nuclear masks. A dataset of images from over 1600 patients was used to train, validate, and test our algorithm; data from 20% of patients were reserved for testing. An external evaluation set with images from 508 patients was used to further validate our findings. The proposed method consistently outperformed other state-of-the art architectures achieving a test per patient area under the receiver operating characteristic curve (AUC-ROC) of 0.87. Performance was comparable to expert colposcopy with a test sensitivity and specificity of 0.94 (p = 0.3) and 0.58 (p = 1.0), respectively. Patients with recurrent human papillomavirus (HPV) infections are at a higher risk of developing cervical cancer. Thus, we sought to incorporate HPV DNA test results as a feature to inform prediction. We found that incorporating patient HPV status improved test specificity to 0.71 at a sensitivity of 0.94.
AB - Cervical cancer is a public health emergency in low- and middle-income countries where resource limitations hamper standard-of-care prevention strategies. The high-resolution endomicroscope (HRME) is a low-cost, point-of-care device with which care providers can image the nuclear morphology of cervical lesions. Here, we propose a deep learning framework to diagnose cervical intraepithelial neoplasia grade 2 or more severe from HRME images. The proposed multi-task convolutional neural network uses nuclear segmentation to learn a diagnostically relevant representation. Nuclear segmentation was trained via proxy labels to circumvent the need for expensive, manually annotated nuclear masks. A dataset of images from over 1600 patients was used to train, validate, and test our algorithm; data from 20% of patients were reserved for testing. An external evaluation set with images from 508 patients was used to further validate our findings. The proposed method consistently outperformed other state-of-the art architectures achieving a test per patient area under the receiver operating characteristic curve (AUC-ROC) of 0.87. Performance was comparable to expert colposcopy with a test sensitivity and specificity of 0.94 (p = 0.3) and 0.58 (p = 1.0), respectively. Patients with recurrent human papillomavirus (HPV) infections are at a higher risk of developing cervical cancer. Thus, we sought to incorporate HPV DNA test results as a feature to inform prediction. We found that incorporating patient HPV status improved test specificity to 0.71 at a sensitivity of 0.94.
KW - Cervical precancer
KW - Endomicroscopy
KW - Multi-task learning
KW - Point-of-care
KW - Neural Networks, Computer
KW - Humans
KW - Colposcopy/methods
KW - Early Detection of Cancer/methods
KW - Pregnancy
KW - Uterine Cervical Dysplasia/diagnostic imaging
KW - Papillomavirus Infections/diagnostic imaging
KW - Uterine Cervical Neoplasms/diagnostic imaging
KW - Sensitivity and Specificity
KW - Female
UR - http://www.scopus.com/inward/record.url?scp=85126307317&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85126307317&partnerID=8YFLogxK
U2 - 10.1016/j.compmedimag.2022.102052
DO - 10.1016/j.compmedimag.2022.102052
M3 - Article
C2 - 35299096
AN - SCOPUS:85126307317
SN - 0895-6111
VL - 97
SP - 102052
JO - Computerized Medical Imaging and Graphics
JF - Computerized Medical Imaging and Graphics
M1 - 102052
ER -