Find research datasets worth reusing

Facial Genetic Syndromes Database

<p>This database contains 3544 face images of individuals with 11 genetic conditions and unaffected (without any diagnosed condition). As well as raw images and syndrome labels, we provide face bounding boxes, 5-point facial landmarks on the entire database, and Human Phenotype Ontology (HPO) annotations of 171 test images. As described in the paper, all images used in this study were identified through searches of publicly available websites and were used for noncommercial research purposes.&nbsp; &nbsp;</p> <p>We cannot share the original images and related data with requestors but have provided links to the URLs in the manuscript describing this work and snapshots in lower resolution,&nbsp;with the assumption that these would not be used for purposes that would not be considered fair use. These data were compiled to produce a new, derivative work, which we offer as a whole. We cannot guarantee that the URLs or images are accurate or up-to-date and encourage interested parties to refer to the sources.</p> <p>For any use of the images in the database, please respect the license and copyrights of the original images and either do not use any visual material or make sure you have acquired rights to use the visual material.&nbsp;&nbsp;</p> <p><br> <strong>Reference Conference Paper:</strong><br> &Ouml;mer S&uuml;mer, Rebekah L. Waikel, Suzanna E. Ledgister Hanchard, Dat Duong, Peter Krawitz, Cristina Conati, Benjamin D. Solomon, Elisabeth Andr&eacute;, &quot;Region-based Saliency Explanations on the Recognition of Facial Genetic Syndromes,&quot; Proceedings of the 8th Machine Learning for Healthcare Conference, PMLR, 2023.</p> <p><strong>Abstract:</strong><br> Deep neural networks in computer vision have shown remarkable progress in recognizing facial genetic syndromes. Many genetic syndromes are difficult to detect, even for experienced clinicians, and computer-aided phenotyping can accelerate clinical diagnosis. High-stakes clinical tasks using deep learning, as in clinical genetics, require human understandable explanations for model decisions. Saliency methods are used to explain DNN predictions in various image analysis domains but have yet to be studied in facial genetics. The syndromic features of most genetic conditions are often localized to areas like the eyes, nose, and mouth. In this paper, to summarize the contribution of key facial regions to a specific disease prediction, we propose a face region relevance score that can be applied to any saliency method. We also investigate how prior knowledge, namely human phenotype ontology and DNN model explanations, align. Quantitative experiments are performed on a new database containing over 3,500 images of 11 rare facial syndromes, a healthy control group, and an additional test set of 171 facial images, whose respective facial phenotypes are labeled by clinicians. Current saliency methods are good at capturing dysmorphism in particular regions (parts of the face), but they may not completely capture all the relevant features in a given person or condition. Our study indicates which saliency explanations and face regions are more consistent with the phenotypes of specific genetic syndromes and could be used in large-scale clinical evaluations.</p>