Accurate Detection of Breaking Boundaries in Tissue Damage

Aim: This study compares logistic regression and CNNs for detecting tissue damage boundaries in medical imaging. Materials and Methods: Group 1 refers to the Logistic Regression with 26 samples and Group 2 refers to the Convolutional Neural Network (CNN). Result: The CNN performs higher gain than Logistic Regression in identifying boundary breaking in tissue damage. The detection effectiveness of CNN is 98.98%, while that of Logistic Regression is 92.19%. Group 2 performed better than Group 1 in detection effectiveness, with the highest detection effectiveness of 98.98% and a significance value of 0.0056 for identifying boundary breaking. Conclusion: In this research, CNN identifies breaking boundaries in tissue damage with a much higher accuracy than Logistic Regression.