Bridging the Gap in MS Diagnosis: Predictive Modelling of Disease Progression Using Logistic Regression and KNN

Multiple sclerosis which is a common demyelinating disease of the central nervous system results in a range of manifestations including visual impairment, muscle weakness and ataxia. There is also much truth to the fact that good treatment and control stem from accurate and fast diagnosis. In this paper, Logistic Regression and KNN classification models are evaluated for detecting MS, employing a dataset with many clinical and MRI features. From the dataset, all the classification basis seems to be provided which include age, gender, early symptoms as well as MRI results. Concerning to the precision, the recall and the F1-score, the logistic regression exhibited higher results for both groups compared to the KNN method. For Class 1, logistic regression results in particular achieved precise measurement of 0.75 and recall of 0.88; for Class 2, a precise measurement of 0.89 and total accuracy of 0.82 were observed. KNN yielded average our accuracy of 0.78 tp for precision, recall, and F1-scores which was 0.71 and 0.83, 0.77 & 0.85 respectively for Class 1 & Class 2 with an overall accuracy of 0.74 for Logistic Regression These findings make Logistic Regression a better model for MS detection and therefore provides a significant tool for early diagnosis and maybe improving outcomes.