Machine learning approach in mortality rate prediction for hemodialysis patients.

Kernel support vector machine algorithm and K-means clustering algorithm are used to determine the expected mortality rate for hemodialysis patients. The national nephrology database of Montenegro has been used to conduct this research. Mortality rate prediction is realized with accuracy up to 94.12% and up to 96.77%, when a complete database is observed and when a reduced database (that contains data for the three most common basic diseases) is observed, respectively. Additionally, it is shown that just a few parameters, most of which are collected during the sole patient examination, are enough for satisfying results.

Medicolegal characteristics of firearm homicides in Belgrade, Serbia: before, during, and after the war in the Former Yugoslavia.

A comparative analysis of firearm homicides committed in Belgrade was performed including four representative years: 1987 (before the civil war in the Former Yugoslavia), 1991 (beginning of the war), 1997 (end of the war), and 2007 (period of social stabilization). The increase in the number of homicides was established in 1991 and 1997 compared with 1987, with the decrease in 2007, but with the continuous increase in the percentage of firearm homicides in the total number of homicides, from 12% in 1987 up to 56% in 2007. The significant increase in firearm homicides during the last decade of the 20th century can be explained by the social disturbances and the high availability of firearms, while their reduction in 2007 could be linked to the gradual stabilization of social circumstances. The results showed that the actual social, political, and economical changes strongly influenced medicolegal characteristics of homicides and particularly firearm homicides.