Conducting remote medical asylum evaluations in the United States during COVID-19: Clinicians' perspectives on acceptability, challenges and opportunities

Utilizing a participatory curriculum development approach for multidisciplinary training on the forensic medical evaluation of asylum seekers

BACKGROUND

Clinicians play an important role in asylum applications through the forensic medical evaluation (FME). The lack of adequately trained and knowledgeable clinicians limits access to FME. Participatory curriculum development is a powerful tool that elevates voices of multiple stakeholders to generate innovation in FME education. The objective of this study was to conduct an interview-based curricular needs assessment of the core skills needed to perform safe and effective FME and the most effective teaching methods targeting multidisciplinary learners.

METHODS

In accordance with a participatory curriculum development framework, we conducted semi-structured interviews of individuals in four key stakeholder groups that play an important role in FME: asylees, experienced educators, prospective learners, and attorneys. We used grounded theory, an inductive approach to the thematic coding of interview transcripts.

RESULTS

Interview participants described the most important skills for performing FME and approaches to teaching these skills. Thematic saturation was reached at 13 interviews. Four major themes central to an FME curriculum were identified: (1) Core knowledge and technical skills to perform effective FME, (2) Practical skills in a trauma-informed approach to FME, (3) Mitigating secondary trauma and building resilience, and (4) Teaching approaches for multi-disciplinary learners.

CONCLUSION

We conducted an interview-based study utilizing participatory curriculum development principles to investigate the most important skills to conduct safe and effective FME of asylum seekers. We found that experiential training that emphasizes the practice of skills in a multi-disciplinary environment is more aligned with stakeholder needs than existing frameworks built around one-way knowledge transfer.

Physical assessment of tortures: Interobserver analysis of medico-legal degrees of consistency based on the Istanbul protocol

The Istanbul Protocol section on medicolegal evaluation delineates the guidelines for the correct management of the physical examination and the methods of assigning the degrees of consistency. Considering that most cases exhibit highly heterogeneous lesions, the examiner is forced to rely on his own experience, and their evaluation may be very subjective. The purpose of this work is to understand how subjective such an evaluation may be and whether the "experience" factor, interpreted as years dedicated to this profession and the number of cases evaluated, is statistically significant. To this end, a survey containing eleven cases of pre-evaluated asylum seekers was sent to thirty Italian clinical forensic practitioners. The participants were invited to assign a degree of consistency to each case according to the Istanbul Protocol guidelines, besides answering a few questions regarding their professional record. The doctors were divided into groups based on the number of cases evaluated and the experience collected expressed in years, and then interobserver analysis was performed. Results showed that the Fleiss' Kappa coefficient acquired significant values when attention was turned to the sub-samples composed of more experienced participants. Therefore, the introduction of appropriately trained health professionals - "experts in migrations and torture" - could lower the risks of misinterpretation and make the assessment as reproducible as possible.

Visual Object Detection with DETR to Support Video-Diagnosis Using Conference Tools

Real-time multilingual phrase detection from/during online video presentations—to support instant remote diagnostics—requires near real-time visual (textual) object detection and preprocessing for further analysis. Connecting remote specialists and sharing specific ideas is most effective using the native language. The main objective of this paper is to analyze and propose—through DEtection TRansformer (DETR) models, architectures, hyperparameters—recommendation, and specific procedures with simplified methods to achieve reasonable accuracy to support real-time textual object detection for further analysis. The development of real-time video conference translation based on artificial intelligence supported solutions has a relevant impact in the health sector, especially on clinical practice via better video consultation (VC) or remote diagnosis. The importance of this development was augmented by the COVID-19 pandemic. The challenge of this topic is connected to the variety of languages and dialects that the involved specialists speak and that usually needs human translator proxies which can be substituted by AI-enabled technological pipelines. The sensitivity of visual textual element localization is directly connected to complexity, quality, and the variety of collected training data sets. In this research, we investigated the DETR model with several variations. The research highlights the differences of the most prominent real-time object detectors: YOLO4, DETR, and Detectron2, and brings AI-based novelty to collaborative solutions combined with OCR. The performance of the procedures was evaluated through two research phases: a 248/512 (Phase1/Phase2) record train data set, with a 55/110 set of validated data instances for 7/10 application categories and 3/3 object categories, using the same object categories for annotation. The achieved score breaks the expected values in terms of visual text detection scope, giving high detection accuracy of textual data, the mean average precision ranging from 0.4 to 0.65.