Background and Aims One of the most important complications of heart transplantation is organ rejection, which is diagnosed on endomyocardial biopsies by pathologists. Computer-based systems could assist in the diagnostic process and potentially improve reproducibility. Here, we evaluated the feasibility of using deep learning in predicting the degree of cellular rejection from pathology slides as defined by the International Society for Heart and Lung Transplantation (ISHLT) grading system. Methods We collected 1079 histopathology slides from 325 patients from three transplant centers in Germany. We trained an attention-based deep neural network to predict rejection in the primary cohort and evaluated its performance using cross validation and by deploying it to three cohorts. Results For binary prediction (rejection yes/no) the mean Area Under the Receiver Operating Curve (AUROC) was 0.849 in the cross-validated experiment and 0.734, 0.729 and 0.716 in external validation cohorts. For a prediction of the ISHLT grade (0R, 1R, 2/3R), AUROCs were 0.835, 0.633 and 0.905 in the cross-validated experiment and 0.764, 0.597, 0.913, and 0.631, 0.633, 0.682, and 0.722, 0.601, 0.805 in the validation cohorts, respectively. The predictions of the AI ​​model were interpretable by human experts and highlighted plausible morphological patterns. Conclusions We conclude that artificial intelligence can detect patterns of cellular transplant rejection in routine pathology, even when trained on small cohorts.

Mitochondria play a crucial role in cell physiology and pathophysiology. In this context, mitochondrial dynamics and, subsequently, mitochondrial ultrastructure have increasingly become hot topics in modern research, with a focus on mitochondrial fission and fusion. Thus, the dynamics of mitochondria in several diseases have been intensively investigated, especially with a view to developing new promising treatment options. However, the majority of recent studies are performed in highly energy-dependent tissues, such as cardiac, hepatic, and neuronal tissues. In contrast, publications on mitochondrial dynamics from the orthopedic or trauma fields are quite rare, even if there are common cellular mechanisms in cardiovascular and bone tissue, especially regarding bone infection. The present report summarizes the spectrum of mitochondrial alterations in the cardiovascular system and compares it to the state of knowledge in the musculoskeletal system. The present paper summarizes recent knowledge regarding mitochondrial dynamics and gives a short, but not exhaustive, overview of its regulation via fission and fusion. Furthermore, the article highlights hypoxia and its accompanying increased mitochondrial fission as a possible link between cardiac ischemia and inflammatory diseases of the bone, such as osteomyelitis. This opens new innovative perspectives not only for the understanding of cellular pathomechanisms in osteomyelitis but also for potential new treatment options.

The pathophysiological role of intracellular bacteria in osteomyelitis is still a matter of debate. Here, we demonstrate for the first time the presence of Staphylococcus aureus internalized into osteoblasts in human tissue samples of a case with a chronic osteomyelitis using ultrastructural transmission electron microscope analysis.

One of the most important complications of heart transplantation is organ rejection, which is diagnosed on endomyocardial biopsies by pathologists. Computer-based systems could assist in the diagnostic process and potentially improve reproducibility. Here, we evaluated the feasibility of using deep learning in predicting the degree of cellular rejection from pathology slides as defined by the International Society for Heart and Lung Transplantation (ISHLT) grading system. We collected 1079 slides from 325 patients from three transplant centers in Germany. We trained an attention-based deep neural network to predict rejection in the primary cohort and evaluated its performance using cross validation and by deploying it to three cohorts. For binary prediction (rejection yes/no) the mean Area Under the Receiver Operating Curve (AUROC) was 0.849 in the cross-validated experiment and 0.734, 0.729 and 0.716 in external validation cohorts. For a prediction of the ISHLT grade (0R, 1R, 2/3R), AUROCs were 0.835, 0.633 and 0.905 in the cross-validated experiment and 0.764, 0.597, 0.913, and 0.631, 0.633, 0.682, and 0.722, 0.601, 0.805 in the validation cohorts, respectively. The predictions of the AI model were interpretable by human experts and highlighted plausible morphological patterns. We conclude that artificial intelligence can detect patterns of cellular transplant rejection in routine pathology, even when trained on small patient cohorts.