Search

How can we predict the occurrence of acute kidney injury (AKI) in cancer patients based on machine learning with serum creatinine data? Given irregular and heterogeneous clinical data, how can we make the most of it for accurate AKI prediction? AKI is a common and significant complication in cancer patients, and correlates with substantial morbidity and mortality. Since no effective treatment for AKI still exists, it is important to take timely preventive measures. While several approaches have been proposed for predicting AKI, their scope and applicability are limited as they either assume regular data measured over a short hospital stay, or do not fully utilize heterogeneous data. In this paper, we provide an AKI prediction model with a greater applicability, which relaxes the constraints of existing approaches, and fully utilizes irregular and heterogeneous data for learning the model. In a cohort of 21,022 cancer patients who were registered into Korea Central Cancer Registry (KCCR) in Seoul National University Hospital between January 1, 2004 and December 31, 2013, our method achieves 0.7892 precision, 0.7506 recall, and 0.7576 F-measure in predicting whether a patient will develop AKI during the next 14 days. [ABSTRACT FROM AUTHOR]

Purpose: To define the diagnostic precision of non-specialized readers in the detection of type 2 endoleaks (T2EL) in arterial versus venous phase acquisitions, and to evaluate an approach for radiation dose reduction. Methods: The pre-discharge and final follow-up multi-slice CT angiographies of 167 patients were retrospectively analyzed. Image data were separated into an arterial and a venous phase reading set. Two radiology residents assessed the reading sets for the presence of a T2EL, feeding vessels, and aneurysm sac size. Findings were compared with a standard of reference established by two experts in interventional radiology. The effective dose was calculated. Results: Overall, experts detected 131 T2ELs, and 331 feeding vessels in 334 examinations. Persistent T2ELs causing aneurysm sac growth > 5 mm were detected in 20 patients. Radiation in arterial and venous phases contributed to a mean of 58.6% and 39.0% of the total effective dose. Findings of reader 1 and 2 showed comparable sensitivities in arterial sets of 80.9 versus 85.5 (p = 0.09), and in venous sets of 73.3 versus 79.4 (p = 0.15), respectively. Reader 1 and 2 achieved a significant higher detection rate of feeding vessels with arterial compared to venous set (p = 0.04, p < 0.01). Both readers correctly identified T2ELs with growing aneurysm sac in all cases, independent of the acquisition phase. Conclusion: Arterial acquisitions enable non-specialized readers an accurate detection of T2ELs, and a significant better identification of feeding vessels. Based on our results, it seems reasonable to eliminate venous phase acquisitions. [ABSTRACT FROM AUTHOR]