Deep learning prediction of renal anomalies for prenatal ultrasound diagnosis.

Autor: Miguel OX; Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada., Kaczmarek E; Children's Hospital of Eastern Ontario Research Institute, Ottawa, Canada., Lee I; Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada., Ducharme R; Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada., Dingwall-Harvey ALJ; Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada.; Children's Hospital of Eastern Ontario Research Institute, Ottawa, Canada., Rennicks White R; Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada.; Department of Obstetrics and Gynecology, University of Ottawa, 501 Smyth Road, Ottawa, ON, K1H-8L6, Canada., Bonin B; Department of Obstetrics and Gynecology, University of Ottawa, 501 Smyth Road, Ottawa, ON, K1H-8L6, Canada.; Department of Obstetrics, Gynecology and Newborn Care, The Ottawa Hospital, Ottawa, Canada., Aviv RI; Department of Radiology and Medical Imaging, University of Ottawa, Ottawa, Canada.; Department of Radiology and Medical Imaging, The Ottawa Hospital, Ottawa, Canada.; Neuroscience Program, Ottawa Hospital Research Institute, Ottawa, Canada., Hawken S; Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada.; Children's Hospital of Eastern Ontario Research Institute, Ottawa, Canada.; School of Epidemiology and Public Health, University of Ottawa, Ottawa, Canada.; ICES, Toronto, Canada., Armour CM; Children's Hospital of Eastern Ontario Research Institute, Ottawa, Canada.; Department of Pediatrics, University of Ottawa, Ottawa, Canada.; Prenatal Screening Ontario, Better Outcomes Registry and Network, Ottawa, Canada., Dick K; Children's Hospital of Eastern Ontario Research Institute, Ottawa, Canada., Walker MC; Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada. mwalker@toh.ca.; Children's Hospital of Eastern Ontario Research Institute, Ottawa, Canada. mwalker@toh.ca.; School of Epidemiology and Public Health, University of Ottawa, Ottawa, Canada. mwalker@toh.ca.; Department of Pediatrics, University of Ottawa, Ottawa, Canada. mwalker@toh.ca.; Department of Obstetrics and Gynecology, University of Ottawa, 501 Smyth Road, Ottawa, ON, K1H-8L6, Canada. mwalker@toh.ca.; International and Global Health Office, University of Ottawa, Ottawa, Canada. mwalker@toh.ca.; BORN Ontario, Children's Hospital of Eastern Ontario, Ottawa, Canada. mwalker@toh.ca.; Department of Obstetrics, Gynecology and Newborn Care, The Ottawa Hospital, Ottawa, Canada. mwalker@toh.ca.
Jazyk: angličtina
Zdroj: Scientific reports [Sci Rep] 2024 Apr 19; Vol. 14 (1), pp. 9013. Date of Electronic Publication: 2024 Apr 19.
DOI: 10.1038/s41598-024-59248-4
Abstrakt: Deep learning algorithms have demonstrated remarkable potential in clinical diagnostics, particularly in the field of medical imaging. In this study, we investigated the application of deep learning models in early detection of fetal kidney anomalies. To provide an enhanced interpretation of those models' predictions, we proposed an adapted two-class representation and developed a multi-class model interpretation approach for problems with more than two labels and variable hierarchical grouping of labels. Additionally, we employed the explainable AI (XAI) visualization tools Grad-CAM and HiResCAM, to gain insights into model predictions and identify reasons for misclassifications. The study dataset consisted of 969 ultrasound images from unique patients; 646 control images and 323 cases of kidney anomalies, including 259 cases of unilateral urinary tract dilation and 64 cases of unilateral multicystic dysplastic kidney. The best performing model achieved a cross-validated area under the ROC curve of 91.28% ± 0.52%, with an overall accuracy of 84.03% ± 0.76%, sensitivity of 77.39% ± 1.99%, and specificity of 87.35% ± 1.28%. Our findings emphasize the potential of deep learning models in predicting kidney anomalies from limited prenatal ultrasound imagery. The proposed adaptations in model representation and interpretation represent a novel solution to multi-class prediction problems.
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje