Decoding medical educators' perceptions on generative artificial intelligence in medical education.

Generative AI (GenAI) is a disruptive technology likely to generate a major impact on faculty and learners in medical education. This work aims to measure the perception of GenAI among medical educators and to gain insights into its major advantages and concerns in medical education. A survey invitation was distributed to medical education faculty of colleges of allopathic and osteopathic medicine within a single university during the fall of 2023. The survey comprised 12 items, among those assessing the role of GenAI for students and educators, the need to modify teaching approaches, GenAI's perceived advantages, applications of GenAI in the educational context, and the concerns, challenges, and trustworthiness associated with GenAI. Responses were obtained from 48 faculty. They showed a positive attitude toward GenAI and disagreed on GenAI having a very negative effect on either the students' or faculty's educational experience. Eighty-five percent of our medical schools' faculty responded to had heard about GenAI, while 42% had not used it at all. Generating text (33%), automating repetitive tasks (19%), and creating multimedia content (17%) were some of the common utilizations of GenAI by school faculty. The majority agreed that GenAI is likely to change its role as an educator. A perceived advantage of GenAI in conducting more effective background research was reported by 54% of faculty. The greatest perceived strengths of GenAI were the ability to conduct more efficient research, task automation, and increased content accessibility. The faculty's major concerns were cheating in home assignments in assessment (97%), tendency for blunder and false information (95%), lack of context (86%), and removal of human interaction in important feedback processes (83%). The majority of the faculty agrees on the lack of guidelines for safe use of GenAI from both a governmental and an institutional policy. The main perceived challenges were cheating, the tendency of GenAI to make errors, and privacy concerns.The faculty recognized the potential impact of GenAI in medical education. Careful deliberation of the pros and cons of GenAI is needed for its effective integration into medical education. There is general agreement that plagiarism and lack of regulations are two major areas of concern. Consensus-based guidelines at the institutional and/or national level need to start to be implemented to govern the appropriate use of GenAI while maintaining ethics and transparency. Faculty responses reflect an optimistic and favorable outlook on GenAI's impact on student learning.

On the role of REM sleep microstructure in suppressing interictal spikes in Electrical Status Epilepticus during Sleep.

OBJECTIVE: Non-Rapid Eye Movement (NREM) sleep promotes the spread and propagation of Interictal Epileptiform Discharges (IEDs), while IEDs are suppressed during REM. Recently, it has been shown that the inhibitory effect on epileptic activity is mostly exerted by the phasic REM (PREM) microstate. This study aims at assessing if this holds true even in the extreme condition of IEDs activation during sleep represented by Electrical Status Epilepticus during Sleep (ESES). METHODS: eight patients affected by ESES, who underwent long-term EEG were included. REM was subdivided into phasic and tonic microstates along with the sleep scoring. IEDs count was carried out using a semi-automatic method and a Spike Index (SI) was calculated. RESULTS: The SI was significantly higher in NREM sleep than in REM. Within REM, the SI was significantly lower in PREM than in tonic REM (TREM). The SI was reduced by 84% in TREM with respect to NREM and by 97% in PREM with respect to NREM. Moreover, the SI was reduced by 87% in PREM with respect to TREM. CONCLUSIONS: PREM has a greater suppressive effect on epileptic activity even in the extreme IEDs activation during sleep typical of ESES. SIGNIFICANCE: Understanding the protective effect of PREM sleep on epileptic activity might be relevant for future therapeutic approaches.

The Data Efficiency of Deep Learning Is Degraded by Unnecessary Input Dimensions.

Biological learning systems are outstanding in their ability to learn from limited training data compared to the most successful learning machines, i.e., Deep Neural Networks (DNNs). What are the key aspects that underlie this data efficiency gap is an unresolved question at the core of biological and artificial intelligence. We hypothesize that one important aspect is that biological systems rely on mechanisms such as foveations in order to reduce unnecessary input dimensions for the task at hand, e.g., background in object recognition, while state-of-the-art DNNs do not. Datasets to train DNNs often contain such unnecessary input dimensions, and these lead to more trainable parameters. Yet, it is not clear whether this affects the DNNs' data efficiency because DNNs are robust to increasing the number of parameters in the hidden layers, and it is uncertain whether this holds true for the input layer. In this paper, we investigate the impact of unnecessary input dimensions on the DNNs data efficiency, namely, the amount of examples needed to achieve certain generalization performance. Our results show that unnecessary input dimensions that are task-unrelated substantially degrade data efficiency. This highlights the need for mechanisms that remove task-unrelated dimensions, such as foveation for image classification, in order to enable data efficiency gains.

[Extremity vascular traumas]. / Traumi vascolari degli arti.

OBJECTIVE: The Authors report on a series of 61 vascular traumas treated over a 7 years, separated in two groups. The first one includes 35 cases, that are street accidents, on the work and gunshot wounds. The second group includes 26 iatrogenic causes due to arterial catheterism. METHODS: All patients underwent ecocolor Doppler directly in the operating theatre and, when this diagnostic procedure was not enough, pre-operating angiography was used (10 cases of complex traumas of the lower limb). RESULTS: One death was reported far each groups (3.27%). In 55 cases (90.1%), limb savage was achieved. In the others 4 (6.93%) of the first group, limb demolition was necessary for different causes. In the first group, severe neurological sequelaes were observed in 2 cases and motor deficits caused by tendon lesions in 1 case. CONCLUSION: The good results obtained are the result of the short ischemic interval between the acute event and treatment, thanks to a multidisciplinary approach of a specific equipe, that is rapid as possible.