The College of Science in the Department of Mathematics discussed a seminar entitled (On Using the Persistent Homology to Classify the Data Space of an Artificial Neural Network in the Field of Disease Diagnosis (1)) for postgraduate student Zainab A. Lazem. This seminar presents a method for classifying the data space of an artificial neural network used in disease diagnosis, employing persistent homology. Persistent homology, a tool from topological data analysis, helps identify and extract significant topological features across multiple scales in complex data. Applying this method allows better understanding of the structure within high-dimensional data generated by neural networks, improving classification accuracy and providing insights into the decision-making process of the model in medical diagnosis.
