Machine Learning Applications to Resting-State Functional MR Imaging Analysis.

Machine learning is one of the most exciting and rapidly expanding fields within computer science. Academic and commercial research entities are investing in machine learning methods, especially in personalized medicine via patient-level classification. There is great promise that machine learning methods combined with resting state functional MR imaging will aid in diagnosis of disease and guide potential treatment for conditions thought to be impossible to identify based on imaging alone, such as psychiatric disorders. We discuss machine learning methods and explore recent advances.

A pilot study of implantable cardiac device interrogation by emergency department personnel.

BACKGROUND: Implanted devices (eg, pacemakers and defibrillators) provide valuable information and may be interrogated to obtain diagnostic information and to direct management. During admission to an emergency department (ED), significant time and cost are spent waiting for device manufacturer representatives or cardiologists to access the data. If ED personnel could safely interrogate implanted devices, more rapid disposition could occur, thus leading to potentially better outcomes at a reduced cost. This was a pilot study examining the feasibility of ED device interrogation. METHODS: This was a prospective convenience sample study of patients presenting to the ED with any chief complaint and who had an implantable device capable of being interrogated by a Medtronic reader. After obtaining informed consent, study patients underwent device interrogation by ED research personnel. After reviewing the device data, the physician documented their opinions of the value of data in aiding care. Patients were followed up at intervals ranging from 30 days out to 1 year to determine adverse events relating to interrogation. RESULTS: Forty-four patients underwent device interrogation. Their mean age was 56 ± 14.7 years (range, 28-83), 75% (33/44) were male and 75% (33/44) were hospitalized from the ED. The interrogations took less than 10 minutes 89% of the time. In 60% of the cases, ED physicians reported the data-assisted patient care. No adverse events were reported relating to the ED interrogations. CONCLUSIONS: In this pilot study, we found that ED personnel can safely and quickly interrogate implantable devices to obtain potentially useful clinical data.