BIFOLD Associated Investigator PD Dr. Meyer (DHZB) and Principal Investigator Prof. Dr. Kühne (DHZB, Charité) developed a recurrent neural network (RNN) which is able to predict severe kidney failure better than human professionals. The corresponding paper was published in “Nature Partner Journal (npj) Digital Medicine.”
Kidney failure is a common complication after heart surgery. BIFOLD’s Associated Investigator PD Dr. Alexander Meyer, Principal Investigator Prof. Dr. Titus Kühne and others at German heart Center Berlin (DHZB) developed a recurrent neural network (RNN) that can predict severe kidney failure. A comparison of detection performances of the AI and experienced medical professional showed that the RNN is clearly superior.
“We cannot and do not want to take decisions away from the intensive care physicians. But we want to help them to make the right decision very early on – and perhaps save their patients’ lives in the process.”
PD Dr. Alexander Meyer
The trained computer scientist PD. Dr. Alexander Meyer is in training to become a specialist for heart surgery at the German Heart Institute Berlin and, as Chief Medical Information Officer, is responsible for the development of digital medicine at the DHZB.
A software developed by PD Dr. Alexander Meyer enables the prediction of numerous other postoperative complications and is already being scientifically evaluated in test operations at the DHZB.
More information is available (in German) in the official press release of DHZB.
Authors: Nina Rank, Boris Pfahringer, Jörg Kempfert, Christof Stamm, Titus Kühne, Felix Schoenrath, Volkmar Falk, Carsten Eickhoff & Alexander Meyer
Abstract: Acute kidney injury (AKI) is a major complication after cardiothoracic surgery. Early prediction of AKI could prompt preventive measures, but is challenging in the clinical routine. One important reason is that the amount of postoperative data is too massive and too high-dimensional to be effectively processed by the human operator. We therefore sought to develop a deep-learning-based algorithm that is able to predict postoperative AKI prior to the onset of symptoms and complications. Based on 96 routinely collected parameters we built a recurrent neural network (RNN) for real-time prediction of AKI after cardiothoracic surgery. From the data of 15,564 admissions we constructed a balanced training set (2224 admissions) for the development of the RNN. The model was then evaluated on an independent test set (350 admissions) and yielded an area under curve (AUC) (95% confidence interval) of 0.893 (0.862–0.924). We compared the performance of our model against that of experienced clinicians. The RNN significantly outperformed clinicians (AUC = 0.901 vs. 0.745, p < 0.001) and was overall well calibrated. This was not the case for the physicians, who systematically underestimated the risk (p < 0.001). In conclusion, the RNN was superior to physicians in the prediction of AKI after cardiothoracic surgery. It could potentially be integrated into hospitals’ electronic health records for real-time patient monitoring and may help to detect early AKI and hence modify the treatment in perioperative care.
Publication: Rank, N., Pfahringer, B., Kempfert, J. et al. Deep-learning-based real-time prediction of acute kidney injury outperforms human predictive performance. npj Digit. Med. 3, 139 (2020). https://doi.org/10.1038/s41746-020-00346-8
Authors: Nina Rank, Boris Pfahringer, Jörg Kempfert, Christof Stamm, Titus Kühne, Felix Schoenrath, Volkmar Falk, Carsten Eickhoff & Alexander Meyer
Abstract: Acute kidney injury (AKI) is a major complication after cardiothoracic surgery. Early prediction of AKI could prompt preventive measures, but is challenging in the clinical routine. One important reason is that the amount of postoperative data is too massive and too high-dimensional to be effectively processed by the human operator. We therefore sought to develop a deep-learning-based algorithm that is able to predict postoperative AKI prior to the onset of symptoms and complications. Based on 96 routinely collected parameters we built a recurrent neural network (RNN) for real-time prediction of AKI after cardiothoracic surgery. From the data of 15,564 admissions we constructed a balanced training set (2224 admissions) for the development of the RNN. The model was then evaluated on an independent test set (350 admissions) and yielded an area under curve (AUC) (95% confidence interval) of 0.893 (0.862–0.924). We compared the performance of our model against that of experienced clinicians. The RNN significantly outperformed clinicians (AUC = 0.901 vs. 0.745, p < 0.001) and was overall well calibrated. This was not the case for the physicians, who systematically underestimated the risk (p < 0.001). In conclusion, the RNN was superior to physicians in the prediction of AKI after cardiothoracic surgery. It could potentially be integrated into hospitals’ electronic health records for real-time patient monitoring and may help to detect early AKI and hence modify the treatment in perioperative care.
Publication: Rank, N., Pfahringer, B., Kempfert, J. et al. Deep-learning-based real-time prediction of acute kidney injury outperforms human predictive performance. npj Digit. Med. 3, 139 (2020). https://doi.org/10.1038/s41746-020-00346-8