Publications by authors named "Suzanna Ledgister Hanchard"
Large language models (LLMs) are generating interest in medical settings. For example, LLMs can respond coherently to medical queries by providing plausible differential diagnoses based on clinical notes. However, there are many questions to explore, such as evaluating differences between open- and closed-source LLMs as well as LLM performance on queries from both medical and non-medical users.
View Article and Find Full Text PDFArtificial intelligence (AI) is increasingly used in genomics research and practice, and generative AI has garnered significant recent attention. In clinical applications of generative AI, aspects of the underlying datasets can impact results, and confounders should be studied and mitigated. One example involves the facial expressions of people with genetic conditions.
View Article and Find Full Text PDFArticle Synopsis
- Dysmorphologists face challenges due to the diverse phenotypic variability of human faces, particularly when using Next-Generation Phenotyping (NGP) tools, which are often trained on limited data.
- To address this, the GestaltMatcher Database (GMDB) was created, compiling over 10,980 facial images from various global populations, significantly improving the representation of underrepresented ancestries, especially African and Asian patients.
- The study found that incorporating data from non-European patients enhanced NGP accuracy by over 11% without compromising performance for European patients, highlighting the importance of diverse datasets in identifying genetic disorders.
Importance: The lack of standardized genetics training in pediatrics residencies, along with a shortage of medical geneticists, necessitates innovative educational approaches.
Objective: To compare pediatric resident recognition of Kabuki syndrome (KS) and Noonan syndrome (NS) after 1 of 4 educational interventions, including generative artificial intelligence (AI) methods.
Design, Setting, And Participants: This comparative effectiveness study used generative AI to create images of children with KS and NS.
Article Synopsis
- AI for facial diagnostics is being used in genetics clinics to identify potential genetic conditions, primarily through Deep Learning (DL) technologies, which show high accuracy for many disorders.
- A study comparing eye-tracking of geneticists and non-clinicians revealed significant differences in how humans and DL models focus on images of individuals with genetic conditions, with notable discrepancies in visual attention patterns.
- The findings suggest that better understanding of these differences can enhance the development and implementation of AI tools in clinical settings, fostering improved integration between clinicians and AI technologies.
Artificial intelligence (AI) is used in an increasing number of areas, with recent interest in generative AI, such as using ChatGPT to generate programming code or DALL-E to make illustrations. We describe the use of generative AI in medical education. Specifically, we sought to determine whether generative AI could help train pediatric residents to better recognize genetic conditions.
View Article and Find Full Text PDFArticle Synopsis
- - Deep learning and AI are being applied in medical genetics, particularly for diagnosing potential genetic conditions using image evaluations.
- - A study was conducted to compare how geneticist clinicians and non-clinicians visually assess these images, using eye-tracking analyses and DL-based saliency maps for comparison.
- - Results indicated significant differences in visual attention between humans and the DL model, with clinicians and non-clinicians showing distinct patterns in their image inspection.
Article Synopsis
- The significant phenotypic variability of human faces complicates the work of dysmorphologists by challenging Next-Generation Phenotyping (NGP) tools, especially when analyzing patients from diverse genetic backgrounds.
- The research established the GestaltMatcher Database (GMDB), which includes over 10,000 facial images from patients with rare genetic disorders worldwide, striving to improve representation of underrepresented populations, particularly Asian and African patients.
- The analysis showed that incorporating data from non-European patients enhanced the accuracy of NGP in diagnosing facial disorders without negatively affecting performance on European patients, emphasizing the need for more diverse datasets in medical genetics.
Deep learning (DL) is applied in many biomedical areas. We performed a scoping review on DL in medical genetics. We first assessed 14,002 articles, of which 133 involved DL in medical genetics.
View Article and Find Full Text PDF