Background Explainability is one of the most heavily debated topics when it comes to the application of artificial intelligence (AI) in healthcare. Even though AI-driven systems have been shown to outperform humans in certain analytical tasks, the lack of explainability continues to spark criticism. Yet, explainability is not a purely technological issue, instead it invokes a host of medical, legal, ethical, and societal questions that require thorough exploration. This paper provides a comprehensive assessment of the role of explainability in medical AI and makes an ethical evaluation of what explainability means for the adoption of AI-driven tools into clinical practice. Methods Taking AI-based clinical decision support systems as a case in point, we adopted a multidisciplinary approach to analyze the relevance of explainability for medical AI from the technological, legal, medical, and patient perspectives. Drawing on the findings of this conceptual analysis, we then conducted an ethical assessment using the “Principles of Biomedical Ethics” by Beauchamp and Childress (autonomy, beneficence, nonmaleficence, and justice) as an analytical framework to determine the need for explainability in medical AI. Results Each of the domains highlights a different set of core considerations and values that are relevant for understanding the role of explainability in clinical practice. From the technological point of view, explainability has to be considered both in terms how it can be achieved and what is beneficial from a development perspective. When looking at the legal perspective we identified informed consent, certification and approval as medical devices, and liability as core touchpoints for explainability. Both the medical and patient perspectives emphasize the importance of considering the interplay between human actors and medical AI. We conclude that omitting explainability in clinical decision support systems poses a threat to core ethical values in medicine and may have detrimental consequences for individual and public health. Conclusions To ensure that medical AI lives up to its promises, there is a need to sensitize developers, healthcare professionals, and legislators to the challenges and limitations of opaque algorithms in medical AI and to foster multidisciplinary collaboration moving forward.

Brain vessel status is a promising biomarker for better prevention and treatment in cerebrovascular disease. However, classic rule-based vessel segmentation algorithms need to be hand-crafted and are insufficiently validated. A specialized deep learning method—the U-net—is a promising alternative. Using labeled data from 66 patients with cerebrovascular disease, the U-net framework was optimized and evaluated with three metrics: Dice coefficient, 95% Hausdorff distance (95HD) and average Hausdorff distance (AVD). The model performance was compared with the traditional segmentation method of graph-cuts. Training and reconstruction was performed using 2D patches. A full and a reduced architecture with less parameters were trained. We performed both quantitative and qualitative analyses. The U-net models yielded high performance for both the full and the reduced architecture: A Dice value of ~0.88, a 95HD of ~47 voxels and an AVD of ~0.4 voxels. The visual analysis revealed excellent performance in large vessels and sufficient performance in small vessels. Pathologies like cortical laminar necrosis and a rete mirabile led to limited segmentation performance in few patients. The U-net outperfomed the traditional graph-cuts method (Dice ~0.76, 95HD ~59, AVD ~1.97). Our work highly encourages the development of clinically applicable segmentation tools based on deep learning. Future works should focus on improved segmentation of small vessels and methodologies to deal with specific pathologies.

(deutsch)Einleitung Abstract (English)The concept of cortical-subcortical loops emphasizes the importance of the basal ganglia for motor, psychomotor, and emotional cortical functions. These loops are bidirectionally controlled by the midbrain dopaminergic system, predominantly but not exclusively at the level of the striatum, including the accumbens nucleus. Successful behaviors increase the activities of the mesostriatal (arising in the complex part of the substantia nigra) and mesolimbic (arising in the ventral tegmental area, VTA) neurons, thereby reinforcing the corresponding actions. In contrast, unsuccessful behaviors result in an increased activation of the lateral habenular complex (LHb), thereby decreasing the activities of mesolimbic neurons. Correspondingly, electrical stimulation of the LHb effectively blocks neuronal activity in the VTA. Whether this block is due to an inhibitory projection from the LHb to the VTA, or whether axons from excitatory LHb neurons target inhibitory neurons within the VTA, is currently not known.Here we show, using in-situ-hybridization and immunocytochemical double labelling at the light and electron microscopic level, that GABAergic neurons are scarce in the LHb and that glutamatergic axons from the LHb mostly target GABAergic neurons in the VTA and the mesopontine rostromedial tegmental nucleus. These data explain the inhibitory effect of LHb activation on the VTA. In addition, however, a small number of LHb terminals in the VTA actually contacts dopaminergic neurons. The biological importance of these terminals requires further investigation.

he prospect of improved clinical outcomes and more efficient health systems has fueled a rapid rise in the development and evaluation of AI systems over the last decade. Because most AI systems within healthcare are complex interventions designed as clinical decision support systems, rather than autonomous agents, the interactions among the AI systems, their users and the implementation environments are defining components of the AI interventions' overall potential effectiveness. Therefore, bringing AI systems from mathematical performance to clinical utility needs an adapted, stepwise implementation and evaluation pathway, addressing the complexity of this collaboration between two independent forms of intelligence, beyond measures of effectiveness alone 1 . Despite indications that some AI-based algorithms now match the accuracy of human experts within preclinical in silico studies 2 , there