Recent advances in Artificial Intelligence have led to some speculation that AI might one day replace medical doctors. Although we tend to overestimate short-term effects, AI is undoubtedly one of the most promising technologies of our times. On the other hand, diagnosis is a very complicated process and, until now, deep learning networks have suffered from technical vulnerabilities and a lack of validation and explanatory capacity.
To quote Yuval Harari, philosophers are very patient people, but engineers are far less patient, and investors are the least patient of all. We strongly believe that several stakeholders of the healthcare system, including those with a complete understanding of the intricacies of medical diagnostics and those who hold patient care paramount, need to shape the conversation on the future of clinical imaging. Understanding AI algorithms and its limitations will be important for radiologists, pathologists, and all other medical specialists engaged with image-based diagnosis.