While past research has emphasized the importance of late blight infection detection and classification, anticipating the potato late blight infection is crucial from the economic point of view as it helps to significantly reduce the production cost. Furthermore, it is necessary to minimize the exposure of potatoes to harmful chemicals and pesticides due to their potential adverse effects on the human immune system. Our work is based on the precise classification of late blight infections in potatoes in European countries using real-time data from 1980 to 2000. To predict the potato late blight outbreak, we incorporated several hybrid machine learning models, as well as a unique combination of stacking classifier and logistic regression, achieving the highest prediction accuracy of 87.22%. Further enhancements of these models and the use of new data sources may lead to a higher late blight prediction accuracy and, consequently, a higher efficiency in managing potatoes’ health.

Physical therapy is a critical component of medical rehabilitation, aiding recovery from conditions such as stroke, spinal cord injuries, and musculoskeletal disorders. Effective rehabilitation requires precise monitoring of patient performance to ensure exercises are executed correctly and progress is accurately assessed. This paper presents a novel automated system for controlling the rehabilitation process and evaluating physical therapy exercise quality using computer vision and a customized Human Skeleton-based Balanced Time Warping algorithm. The proposed method quantitatively assesses the similarity between a physiotherapist and patient performance by analyzing skeletal motion data extracted from RGB-D video sequences without requiring pre-alignment or sensor-specific calibration. A motion-dependent, weighted Euclidean distance between 3D skeletal models is used to compute pose dissimilarity, while a modified DTW approach aligns temporal sequences and evaluates dynamic consistency. The total dissimilarity measure is a balanced combination of posture (DP) and dynamics (DT) components. Evaluated on a custom dataset of 136 video recordings from 23 participants performing exercises in sitting and standing positions under varying performance accuracy levels (“good,” “intermediate,” and “bad”), the system demonstrates the strong clustering of accuracy levels. Proposed dissimilarity, together with a fixed reference element (physiotherapist), induces a natural non-strict order on the set of distances between patients and physiotherapists. A high value of Spearman’s rank correlation coefficient between computed dissimilarity and execution accuracy (0.977) indicates that this method is suitable for assessing exercise performance accuracy and for adequately evaluating the patient’s rehabilitation progress. The method enables objective, real-time feedback, reduces therapist workload, and supports remote monitoring, offering a scalable solution for personalized rehabilitation. Future work will involve clinical validation with post-stroke and cardiac patients.

The paper focuses on developing a method for evaluating the similarity between skeletal models of an instructor and a patient during physical therapy exercises. Unlike general-purpose measures, our approach considers the specific characteristics of therapeutic exercises, including initial positioning, predominant movements, and exercise pace. By incorporating the concept of informativeness for skeletal points and refining normalization techniques, we improve the accuracy of pairwise dissimilarity measures. Experimental results demonstrate improved separability of records based on the accuracy of exercise repetition, highlighting the potential of this approach for enhancing automated physical rehabilitation systems.