Does improving diagnostic accuracy increase artificial intelligence adoption? A public acceptance survey using randomized scenarios of diagnostic methods - Ecole Centrale de Marseille
Article Dans Une Revue Artificial Intelligence in Health Année : 2024

Does improving diagnostic accuracy increase artificial intelligence adoption? A public acceptance survey using randomized scenarios of diagnostic methods

Résumé

This study examines the acceptance of artificial intelligence (AI)-based diagnostic alternatives compared to traditional biological testing through a randomized scenario experiment in the domain of neurodegenerative diseases (NDs). A total of 3225 pairwise choices of ND risk-prediction tools were offered to participants, with 1482 choices comparing AI with the biological saliva test and 1743 comparing AI+ with the saliva test (with AI+ using digital consumer data, in addition to electronic medical data). Overall, only 36.68% of responses showed preferences for AI/AI+ alternatives. Stratified by AI sensitivity levels, acceptance rates for AI/AI+ were 35.04% at 60% sensitivity and 31.63% at 70% sensitivity, and increased markedly to 48.68% at 95% sensitivity (p <0.01). Similarly, acceptance rates by specificity were 29.68%, 28.18%, and 44.24% at 60%, 70%, and 95% specificity, respectively (P < 0.01). Notably, AI consistently garnered higher acceptance rates (45.82%) than AI+ (28.92%) at comparable sensitivity and specificity levels, except at 60% sensitivity, where no significant difference was observed. These results highlight the nuanced preferences for AI diagnostics, with higher sensitivity and specificity significantly driving acceptance of AI diagnostics.
Fichier principal
Vignette du fichier
manuscript_aih01923.pdf (1008.1 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04746007 , version 1 (21-10-2024)

Licence

Identifiants

Citer

Yulin Hswen, Ismaël Rafaï, Antoine Lacombe, Bérengère Davin-Casalena, Dimitri Dubois, et al.. Does improving diagnostic accuracy increase artificial intelligence adoption? A public acceptance survey using randomized scenarios of diagnostic methods. Artificial Intelligence in Health, In press, ⟨10.36922/aih.3561⟩. ⟨hal-04746007⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

More