Linking Symptom Inventories using Semantic Textual Similarity
Abstract
An extensive library of symptom inventories has been developed over time to measure clinical symptoms, but this variety has led to several long standing issues. Most notably, results drawn from different settings and studies are not comparable, which limits reproducibility. Here, we present an artificial intelligence (AI) approach using semantic textual similarity (STS) to link symptoms and scores across previously incongruous symptom inventories. We tested the ability of four pre-trained STS models to screen thousands of symptom description pairs for related content - a challenging task typically requiring expert panels. Models were tasked to predict symptom severity across four different inventories for 6,607 participants drawn from 16 international data sources. The STS approach achieved 74.8% accuracy across five tasks, outperforming other models tested. This work suggests that incorporating contextual, semantic information can assist expert decision-making processes, yielding gains for both general and disease-specific clinical assessment.
Cite
@article{arxiv.2309.04607,
title = {Linking Symptom Inventories using Semantic Textual Similarity},
author = {Eamonn Kennedy and Shashank Vadlamani and Hannah M Lindsey and Kelly S Peterson and Kristen Dams OConnor and Kenton Murray and Ronak Agarwal and Houshang H Amiri and Raeda K Andersen and Talin Babikian and David A Baron and Erin D Bigler and Karen Caeyenberghs and Lisa Delano-Wood and Seth G Disner and Ekaterina Dobryakova and Blessen C Eapen and Rachel M Edelstein and Carrie Esopenko and Helen M Genova and Elbert Geuze and Naomi J Goodrich-Hunsaker and Jordan Grafman and Asta K Haberg and Cooper B Hodges and Kristen R Hoskinson and Elizabeth S Hovenden and Andrei Irimia and Neda Jahanshad and Ruchira M Jha and Finian Keleher and Kimbra Kenney and Inga K Koerte and Spencer W Liebel and Abigail Livny and Marianne Lovstad and Sarah L Martindale and Jeffrey E Max and Andrew R Mayer and Timothy B Meier and Deleene S Menefee and Abdalla Z Mohamed and Stefania Mondello and Martin M Monti and Rajendra A Morey and Virginia Newcombe and Mary R Newsome and Alexander Olsen and Nicholas J Pastorek and Mary Jo Pugh and Adeel Razi and Jacob E Resch and Jared A Rowland and Kelly Russell and Nicholas P Ryan and Randall S Scheibel and Adam T Schmidt and Gershon Spitz and Jaclyn A Stephens and Assaf Tal and Leah D Talbert and Maria Carmela Tartaglia and Brian A Taylor and Sophia I Thomopoulos and Maya Troyanskaya and Eve M Valera and Harm Jan van der Horn and John D Van Horn and Ragini Verma and Benjamin SC Wade and Willian SC Walker and Ashley L Ware and J Kent Werner and Keith Owen Yeates and Ross D Zafonte and Michael M Zeineh and Brandon Zielinski and Paul M Thompson and Frank G Hillary and David F Tate and Elisabeth A Wilde and Emily L Dennis},
journal= {arXiv preprint arXiv:2309.04607},
year = {2023}
}