Safoora Masoumi

Cancer survivors increasingly consult chatbots between visits, but conditions for safe use remain unclear. The goal of this study was to characterize survivors’ judgments of patient-facing LLM responses and derive deployment requirements. We conducted semi-structured interviews with 21 breast and prostate cancer survivors immediately after a structured rating task; transcripts were dual-coded and analyzed thematically. Three themes governed willingness to act: trust, accuracy, and provenance; personalization and empathy in the cancer context; and usability, actionability, and risk-based escalation. Survivors favored citation-first answers, lightweight personalization, plain-language steps, and one-tap clinician escalation. Patient-defined requirements aligned with European priorities, informing trustworthy, integrated survivorship deployments.

Background: Cancer survivorship is a complicated, chronic, and long-lasting experience, causing uncertainty and a wide range of physical and emotional health concerns. Due to the complexity of cancer, patients often seek out multiple sources of health information to better understand the aspects of their cancer diagnosis. The high variability among patients with cancer presents significant challenges in treatment, prognosis, and overall disease management. Artificial intelligence (AI) chatbots can further personalize cancer care delivery. However, there is a knowledge gap regarding cancer survivors’ perceived facilitators and barriers to adopting and using AI chatbots. Objective: In this study, we examined cancer survivors’ experiences of using existing AI chatbots and identified their facilitators and barriers to the adoption of AI chatbots. Methods: We conducted a qualitative study to investigate the perceptions of cancer survivors, conducting semistructured interviews to understand their prior use of existing AI chatbots in general. We asked the participants about their perceptions regarding AI chatbot acceptability and comfort level; trust and adherence; and concerns, barriers, and suggestions. We used the Consolidated Criteria for Reporting Qualitative Research (COREQ) checklist for this qualitative report. Results: Of 21 participants, 17 (81%) were female patients with breast cancer, 15 (71%) were aged 50 to 64 years, 19 (90%) were White, and 9 (43%) had a graduate degree. Participants’ responses were grouped into three overarching themes: (1) patients’ perceptions of interacting with chatbots compared to health care professionals, (2) patient-chatbot interaction, and (3) chatbot information processing. All participants who were interviewed reported that they would prefer interacting with health care professionals over a chatbot. The lack of empathy shown by chatbots was a major concern among cancer survivors. Many patients criticized chatbots for tending to provide a general overarching response to their questions rather than being specific to their cancer diagnosis. The main concerns of cancer survivors with using chatbots were the overabundance of general information that was often not relevant to their diagnosis and privacy of patient information. Conclusions: The findings of this study underscore the critical importance of empathetic responses during AI chatbot interactions for cancer survivors, as the lack of personalized and emotional responses can lead to distrust and frustration. Clinically, these tools should be integrated as supplementary resources to enhance patient engagement while preserving essential human support. Policymakers need to develop guidelines that promote responsible use of AI in cancer care, prioritizing patient confidentiality and trustworthiness. AI chatbots have the potential to significantly improve the support provided to cancer survivors, but it is crucial to address the identified barriers and enhance user acceptance.

A majority (>80%) of maternal deaths in the United States are preventable. Using machine learning (ML) models that are generated from electronic medical records (EMRs) may be a promising approach to predict the risk of adverse maternal outcomes and enable proactive intervention to prevent maternal mortality. Current evidence syntheses of such ML approaches either focus only on specific maternal outcomes, aspects other than risk prediction, or do not consider the full pipeline of studies from the development to implementation in clinical practice. The goal of this scoping review is to document evidence for the use of ML models for predicting the risk of maternal morbidity and mortality outcomes (research objective [RO1]), the translation of such models into applications for clinical use by providers (RO2), and factors associated with the implementation of clinical applications in practice (RO3). The review was limited to studies in health care settings, using data from EMRs. A detailed search string was developed in collaboration with a health sciences librarian and implemented on February 20, 2023, on PubMed, CINAHL Plus, Scopus, Embase, and IEEE Xplore. Two reviewers independently reviewed titles and abstracts for inclusion, and a third reviewer resolved conflicts. Only full-length journal articles published in English were included. Studies using non-EMR data exclusively were excluded. Two reviewers independently reviewed full texts for inclusion, and a third reviewer resolved conflicts. A structured template was used for data extraction, and findings were summarized descriptively. From 480 deduplicated studies identified from the search, 142 studies were included for full-text review, and 39 studies were included in the review. More than half of the included studies were conducted in 2022, and 34 studies were from just 3 countries (United States, China, and Israel). More studies focused on identifying the risk of pregnancy and delivery outcomes compared with postpartum outcomes. The top 3 most common outcomes for risk prediction were cardiovascular risks and hypertensive disorders of pregnancy (9 studies), gestational diabetes (7 studies), and postpartum hemorrhage (6 studies). Data were labeled with computable phenotypes in 30 studies, and the most often used method in ML models was boosting methods (18 studies). The most common metric used to assess model performance was area under the precision-recall curve (AUPRC; 33 studies). No studies described clinical applications of ML models for providers (RO2) or associated implementation factors (RO3). Key recommendations for future research and practice include expanding efforts to study maternal morbidity and mortality outcomes in the postpartum period, increasing transparency and reproducibility of studies through use of reporting checklists, and expanding efforts to implement ML models in clinical practice.

Background: Abstract review is a time and labor-consuming step in the systematic and scoping literature review in medicine. Text mining methods, typically natural language processing (NLP), may efficiently replace manual abstract screening. This study applies NLP to a deliberately selected literature review problem, the trend of using NLP in medical research, to demonstrate the performance of this automated abstract review model. Methods: Scanning PubMed, Embase, PsycINFO, and CINAHL databases, we identified 22,294 with a final selection of 12,817 English abstracts published between 2000 and 2021. We invented a manual classification of medical fields, three variables, i.e., the context of use (COU), text source (TS), and primary research field (PRF). A training dataset was developed after reviewing 485 abstracts. We used a language model called Bidirectional Encoder Representations from Transformers to classify the abstracts. To evaluate the performance of the trained models, we report a micro f1-score and accuracy. Results: The trained models’ micro f1-score for classifying abstracts, into three variables were 77.35% for COU, 76.24% for TS, and 85.64% for PRF. The average annual growth rate (AAGR) of the publications was 20.99% between 2000 and 2020 (72.01 articles (95% CI: 56.80–78.30) yearly increase), with 81.76% of the abstracts published between 2010 and 2020. Studies on neoplasms constituted 27.66% of the entire corpus with an AAGR of 42.41%, followed by studies on mental conditions (AAGR = 39.28%). While electronic health or medical records comprised the highest proportion of text sources (57.12%), omics databases had the highest growth among all text sources with an AAGR of 65.08%. The most common NLP application was clinical decision support (25.45%). Conclusions: BioBERT showed an acceptable performance in the abstract review. If future research shows the high performance of this language model, it can reliably replace manual abstract reviews.

BackgroundAbstract review is a time and labor-consuming step in the systematic and scoping literature review in medicine. Automation methods, typically natural language processing (NLP), may efficiently replace manual abstract screening. This study applies NLP to a deliberately selected literature review problem, the trend of using NLP in medical research, to demonstrate the performance of this automated abstract review model.MethodsScanning PubMed, Embase, PsycINFO, and CINAHL databases, we identified 22,294 with a final selection of 12,817 English abstracts published between 2000 to 2021. We invented a manual classification of medical fields, three variables, i.e., the context of use (COU), text source (TS), and primary research field (PRF). A training dataset was developed after reviewing 485 abstracts. We used a language model called Bidirectional Encoder Representations from Transformers to classify the abstracts. To evaluate the performance of the trained models, we report a micro f1-score and accuracy.ResultsThe trained models' micro f1-score in classifying abstracts the three variables were 77.35 % for COU, 76.24% for TS, and 85.64 % for PRF. The average annual growth rate (AAGR) of the publications was 20.99% between 2000 and 2020 (72.01 articles (95%CI: 56.80-78.30) yearly increase), with 81.76% of the abstracts published between 2010 and 2020. Studies on neoplasms constituted 27.66% of the entire corpus with an AAGR of 42.41%, followed by studies on mental conditions (AAGR=39.28%). While electronic health or medical records comprised the highest proportion of text sources (57.12%), omics databases had the highest growth among all text sources with an AAGR of 65.08%. The most common NLP application was clinical decision support (25.45%).ConclusionsBioBERT showed an acceptable performance in the abstract review. If future research shows the high performance of this language model, it can reliably replace manual abstract reviews.

Background: Meta-analysis is a central method for quality evidence generation. In particular, meta-analysis is gaining speedy momentum in the growing world of quantitative information. There are several software applications to process and output expected results. Open-source software applications generating such results are receiving more attention. This paper uses Python’s capabilities to provide applicable instruction to perform a meta-analysis. Methods: We used the PythonMeta package with several modifications to perform the meta-analysis on an open-access dataset from Cochrane. The analyses were complemented by employing Python’s zEpid package capable of creating forest plots. Also, we developed Python scripts for contour-enhanced funnel plots to assess funnel plots asymmetry. Finally, we ran the analyses in R and STATA to check the cross-validity of the results. Results: A stepwise instruction on installing the software and packages and performing meta-analysis was provided. We shared the Python codes for meta-analysts to follow and generate the standard outputs. Our results were similar to those yielded by R and STATA. Conclusion: We successfully produced standard meta-analytic outputs using Python. This programming language has several flexibilities to improve the meta-analysis results even further.