TY - JOUR AU - Kiptoo, Stephen AU - Nderu, Lawrence AU - Mutanu, Leah PY - 2021/02/20 Y2 - 2024/03/29 TI - Automated Detection of Cervical Pre-Cancerous Lesions Using Regional-Based Convolutional Neural Network JF - International Journal of Sciences: Basic and Applied Research (IJSBAR) JA - IJSBAR VL - 56 IS - 1 SE - Articles DO - UR - https://www.gssrr.org/index.php/JournalOfBasicAndApplied/article/view/12259 SP - 104-123 AB - <p>The Cervical Colposcopy image is an image of woman’s cervix taken with a digital colposcope after application of acetic acid. The captured cervical images must be understood for diagnosis, prognosis and treatment planning of the anomalies. This Cervix image understanding is generally performed by skilled medical professionals. However, the scarcity of human medical experts and the fatigue and rough estimate procedures involved with them limit the effectiveness of image understanding performed by skilled medical professionals. This paper, the model uses Regional Based Convolutional Neural Network (R-CNN) to effectively visualize of pre-cancerous lesions and to aid in diagnosis of the disease. The model was trained, on a dataset comprising of 10,383 cervical images samples. The datasets were derived from public dataset repositories. The training samples comprised of type class 1, 2 and 3 traits of cervical precancerous traits. The performance was evaluated using K-nearest -neighbor model over R-CNN. With an accuracy rate of 86%, this approach heralds a promising development in the detection of cervical precancerous lesions. This study findings established that the proposed model in provision of the better accuracy and misclassifications performance than various testing algorithms.</p> ER -