1272. Deep Learning-Generated Hip Radiographic Measurements Are Fast and Adequate for Reliable Assessment of Hip Dysplasia
Authors * Denotes Presenting Author
  1. Holden Archer *; UT Southwestern
  2. Seth Reine; UT Southwestern
  3. Ahmed Alshaikhsalama; UT Southwestern
  4. Louis Vazquez; UT Southwestern
  5. Ajay Kohli; UT Southwestern
  6. Joel Wells; UT Southwestern
  7. Avneesh Chhabra; UT Southwestern
Hip dysplasia (HD) leads to premature osteoarthritis. Timely detection and correction of HD has been shown to improve pain, functional status, and hip longevity. Several time-consuming radiographic measurements are used to confirm HD. An artificial intelligence (AI) software named HIPPO automatically locates anatomical landmarks on anteroposterior (AP) pelvis radiographs and performs the needed measurements. The primary aim of this study was to assess the reliability of this tool as compared to multireader evaluation in clinically proven cases of adult HD for external validation. The secondary aims were to assess the time savings achieved and evaluate inter-reader assessment.

Materials and Methods:
This study received institutional review board approval. A consecutive preoperative sample of 130 patients with hip dysplasia (82.3% women and 17.7% men with median age of 28.6 years) was used. Three trained readers’ measurements were compared to AI outputs of lateral center edge angle (LCEA), caput-collum-diaphyseal (CCD) angle, pelvic obliquity, Tönnis angle, Sharp’s angle, and femoral head coverage. Intraclass correlation coefficients (ICC) and Bland-Altman analyses were obtained.

Out of 256 hips with AI outputs, all six hip AI measurements were successfully obtained. The AI-reader correlations were generally good (ICC = 0.60 to 0.74) to excellent (ICC> 0.75). There was lower agreement for CCD angle measurement. Most widely used measurements for HD diagnosis (LCEA and Tonnis angle) demonstrated good to excellent inter-method reliability (ICC = 0.71-0.86 and 0.82-0.90). The median reading time for the three readers and AI was 212, 131, 734, and 41 seconds, respectively.

This study validated that the AI-based trained software demonstrated significant time savings in reliable radiographic assessment of patients with hip dysplasia. In addition to providing extensive time savings, integration of this AI system could provide preliminary measurements to physicians and direction for more thorough assessment for HD, especially in places without access to board-certified radiologists or orthopedic surgeons to conduct the measurements.