Characterizing the Progression of Pulmonary Edema Severity: Can Pairwise Comparisons in Radiology Reports Help?

Accurately assessing the severity of pulmonary edema is critical for making treatment decisions in congestive heart failure patients. However, the current scale for quantifying pulmonary edema based on chest radiographs does not have well-characterized severity levels, with substantial inter-radiologist disagreement. In this study, we investigate whether comparisons documented in radiology reports can provide accurate characterizations of pulmonary edema progression. We propose a rules-based natural language processing approach to assess the change in a patient's pulmonary edema status (e.g. better, worse, no change) by performing pairwise comparisons of consecutive radiology reports, using regular expressions and heuristics derived from clinical knowledge. Evaluated against ground-truth labels from expert radiologists, our labeler extracts comparisons describing the progression of pulmonary edema with 0.875 precision and 0.891 recall. We also demonstrate the potential utility of comparison labels in providing additional fine-grained information over noisier labels produced by models that directly estimate severity level.

Knowledge and Attitudes for the Management of Behavioral Variant of Frontotemporal Dementia

Background: The diagnosis of the behavioral variant of frontotemporal dementia (bvFTD) can be especially challenging and is relatively underdiagnosed. There is scarce information on training and attitudes from care providers facing bvFTD in settings with limited resources. We aim to describe clinical knowledge and attitudes facing bvFTD from neurologists, psychiatrists, and residents in Peru.Methods: Potential participants received invitations by email to complete an online questionnaire. In addition, we reviewed 21 curricula from undergraduate medical schools' programs offered by the main schools of medicine in Peru during 2020 and 2021.Results: A total of 145 participants completed the survey. The responders were neurologists (51%), psychiatrists (25%), and residents in neurology or psychiatry (24%). Only 26% of the respondents acknowledged receiving at least one class on bvFTD in undergraduate medical training, but 66.6% received at least some training during postgraduate study. Participants identified isolated supportive symptoms for bvFTD; however, only 25% identified the possible criteria and 18% the probable bvFTD criteria. They identified MoCA in 44% and Frontal Assessment Battery (39%) as the most frequently used screening test to assess bvFTD patients. Memantine and Acetylcholinesterase inhibitors were incorrectly indicated by 40.8% of participants. Seventy six percentage of participants indicated that they did not provide education and support to the caregiver. The dementia topic was available on 95.2%, but FTD in only 19%.Conclusion: Neuropsychiatry medical specialists in Peru receive limited training in FTD. Their clinical attitudes for treating bvFTD require appropriate training focused on diagnostic criteria, assessment tools, and pharmacological and non-pharmacological management.

CPD article: Felid lungworms: from complete strangers to major parasites

The parasitic nematodes Aelurostrongylus abstrusus, Troglostrongylus brevior and Capillaria aerophila affect the respiratory system of cats and are a primary cause of respiratory disease in cats in many countries. While they have been underestimated for a long time, in recent years academics and veterinarians have become more aware of their importance, and now felid lungworms are recognised as primary agents of respiratory disease. Therefore, timely diagnosis and treatment, and efficacious prevention methods are a priority in feline clinical practice. Recent data have unveiled many features of diseases caused by these nematodes, and this article reviews and discusses practical and clinical knowledge, as well as recent updates on clinical management of aelurostrongylosis, troglostrongylosis and capillariosis in cats.

Virtually Transformative Experiences in Geriatric Education: Lessons Pre and Post Pandemic

Georgetown University medical students have the option of selecting a two-week rotation in Geriatrics during their third-year. Since Fall 2019, the curriculum has included three immersive virtual reality (VR) labs: hearing & vision loss, Alzheimer’s disease, and end-of-life conversations created by Embodied Labs. The curricular goals include increasing empathy and sensitivity of learners to the perspective of older adults, decreasing ageism & stereotyping, and increasing clinical knowledge. In each lab, students are immersed in a live film, first-person point of view of an older adult. They interact with the immersive environment via gaze, voice, and natural hand motions. Pre-pandemic, students viewed the labs in-person using a commercial VR headset. Since the pandemic, March 2020, students accessed the VR labs through the virtual modality of Zoom. This abstract summarizes data on knowledge and attitudes examining differences in knowledge and attitudes pre and post-pandemic.

Transforming Education Using Virtual Reality: Geriatrics Clerkship Before and During Pandemic

Since 2006, the Georgetown University School of Medicine has offered a two-week elective in Geriatrics for third-year medical students. Students rotate through diverse clinical experiences, including general geriatrics, geriatric neurology, physical medicine & rehabilitation, memory disorders, Parkinson’s and dementia, and palliative care. In addition, students learn about arts, humanities & ethics, communication skills, and taking the patient’s perspective. In Fall 2019, pre-pandemic, we added virtual reality (VR) experiences focused on hearing & vision loss, Alzheimer’s disease, and end-of-life conversations created by Embodied Labs. Curricular goals included increasing students’ empathy and sensitivity, decreasing ageism & stereotyping, and increasing clinical knowledge. Findings suggest regardless of pandemic (pre vs. during) or modality (in-person vs. Zoom) that after participating in the VR labs, students are slightly more comfortable taking care of older adult patients with dementia as well as hearing & vision loss, and participating in end-of-life conversations.

A class-contrastive human-interpretable machine learning approach to predict mortality in severe mental illness

Machine learning (ML), one aspect of artificial intelligence (AI), involves computer algorithms that train themselves. They have been widely applied in the healthcare domain. However, many trained ML algorithms operate as ‘black boxes’, producing a prediction from input data without a clear explanation of their workings. Non-transparent predictions are of limited utility in many clinical domains, where decisions must be justifiable. Here, we apply class-contrastive counterfactual reasoning to ML to demonstrate how specific changes in inputs lead to different predictions of mortality in people with severe mental illness (SMI), a major public health challenge. We produce predictions accompanied by visual and textual explanations as to how the prediction would have differed given specific changes to the input. We apply it to routinely collected data from a mental health secondary care provider in patients with schizophrenia. Using a data structuring framework informed by clinical knowledge, we captured information on physical health, mental health, and social predisposing factors. We then trained an ML algorithm and other statistical learning techniques to predict the risk of death. The ML algorithm predicted mortality with an area under receiver operating characteristic curve (AUROC) of 0.80 (95% confidence intervals [0.78, 0.82]). We used class-contrastive analysis to produce explanations for the model predictions. We outline the scenarios in which class-contrastive analysis is likely to be successful in producing explanations for model predictions. Our aim is not to advocate for a particular model but show an application of the class-contrastive analysis technique to electronic healthcare record data for a disease of public health significance. In patients with schizophrenia, our work suggests that use or prescription of medications like antidepressants was associated with lower risk of death. Abuse of alcohol/drugs and a diagnosis of delirium were associated with higher risk of death. Our ML models highlight the role of co-morbidities in determining mortality in patients with schizophrenia and the need to manage co-morbidities in these patients. We hope that some of these bio-social factors can be targeted therapeutically by either patient-level or service-level interventions. Our approach combines clinical knowledge, health data, and statistical learning, to make predictions interpretable to clinicians using class-contrastive reasoning. This is a step towards interpretable AI in the management of patients with schizophrenia and potentially other diseases.

Investigating the Effectiveness of Using a Situated Simulation-Based Program to Improve Occupational Therapy Students’ Interactions and Observation Skills with Children

Purpose. Interaction and observation are critical skills for occupational therapists who work with pediatric clients. The objective of this study was to investigate whether using standardized child patients within a situated simulation-based (SSB) program increases students’ knowledge and clinical skills when working with children in occupational therapy. Materials and Methods. This controlled trial with multiple measures recruited students from the pediatric occupational therapy curriculum enrolled in an SSB program in consecutive academic years ( n = 62 ). Experimental group students participated in a simulation experience with video training sessions, followed by an SSB exam with standardized child patients; the control group performed the video training simultaneously. Quantitative outcomes included quizzes to measure clinical knowledge, video training scores, and a situated simulation exam to assess clinical skills. Results. The experimental group had a significantly higher postwritten quiz scores than the control group; the video training scores were not significantly different between groups. Linear regression analysis showed a significant association between the SSB exam and postwritten quiz scores ( β = 0.487 , p = 0.017 ). The experimental group had a total pass rate of 65.6% for the SSB exam. The communication and interaction pass rate was 53.1%; the basic evaluation rate was 68.8%, implying that communication/interaction skills are hard to simulate from video training alone; therefore, the authentic fidelity of the SSB program needs to improve further to enhance learning. Conclusions. The SSB program with standardized child patients improved students’ clinical knowledge and skills more than lectures and practice alone. Using standardized child patients in programs or exams appears to positively influence students’ performance. Situated simulation-based learning that allows the realistic practice of observation and communication skills may enhance students’ clinical competency. Future research should develop standard training methods and evaluation processes in high-fidelity simulations for generalized use in other occupational therapy programs.

FAIR Data Reuse in Traumatic Brain Injury: Exploring Inflammation and Age as Moderators of Recovery in the TRACK-TBI Pilot

The guiding principle for data stewardship dictates that data be FAIR: findable, accessible, interoperable, and reusable. Data reuse allows researchers to probe data that may have been originally collected for other scientific purposes in order to gain novel insights. The current study reuses the Transforming Research and Clinical Knowledge for Traumatic Brain Injury (TRACK-TBI) Pilot dataset to build upon prior findings and ask new scientific questions. Specifically, we have previously used a multivariate analytics approach to multianalyte serum protein data from the TRACK-TBI Pilot dataset to show that an inflammatory ensemble of biomarkers can predict functional outcome at 3 and 6 months post-TBI. We and others have shown that there are quantitative and qualitative changes in inflammation that come with age, but little is known about how this interaction affects recovery from TBI. Here we replicate the prior proteomics findings with improved missing value analyses and non-linear principal component analysis and then expand upon this work to determine whether age moderates the effect of inflammation on recovery. We show that increased age correlates with worse functional recovery on the Glasgow Outcome Scale-Extended (GOS-E) as well as increased inflammatory signature. We then explore the interaction between age and inflammation on recovery, which suggests that inflammation has a more detrimental effect on recovery for older TBI patients.