Enhanced Cellular Detection Using Convolutional Neural Networks and Sliding Window Super-Resolution Inference.

Abstract

Histopathology currently serves as the standard for breast cancer diagnosis, but its manual execution demands time and expertise from pathologists. Artificial intelligence, particularly in digital pathology, has made significant strides, offering new opportunities for precision and efficiency in disease diagnosis. This study presents a methodology to enhance cell nuclei detection in breast cancer histopathological images using convolutional neural network models to apply super-resolution and object detection. Several model architectures are explored, and their performance is evaluated regarding accuracy and sensitivity. The results affirm the potential of the proposed approach for automated cell nuclei identification. These AI advancements in digital pathology open avenues for early and precise cancer detection, influencing clinical practices and patient well-being and improving diagnostic efficiency.