[Study on the risk factors for the misdiagnosis of femoral head osteonecrosis]

Automatic detection of early osteonecrosis of the femoral head from various hip pathologies using deep convolutional neural network: a multi-centre study

PURPOSE

The aim of this study was to develop a deep convolutional neural network (DCNN) for detecting early osteonecrosis of the femoral head (ONFH) from various hip pathologies and evaluate the feasibility of its application.

METHODS

We retrospectively reviewed and annotated hip magnetic resonance imaging (MRI) of ONFH patients from four participated institutions and constructed a multi-centre dataset to develop the DCNN system. The diagnostic performance of the DCNN in the internal and external test datasets was calculated, including area under the receiver operating characteristic curve (AUROC), accuracy, precision, recall, and F1 score, and gradient-weighted class activation mapping (Grad-CAM) technique was used to visualize its decision-making process. In addition, a human-machine comparison trial was performed.

RESULTS

Overall, 11,730 hip MRI segments from 794 participants were used to develop and optimize the DCNN system. The AUROC, accuracy, and precision of the DCNN in internal test dataset were 0.97 (95% CI, 0.93-1.00), 96.6% (95% CI: 93.0-100%), and 97.6% (95% CI: 94.6-100%), and in external test dataset, they were 0.95 (95% CI, 0.91- 0.99), 95.2% (95% CI, 91.1-99.4%), and 95.7% (95% CI, 91.7-99.7%). Compared with attending orthopaedic surgeons, the DCNN showed superior diagnostic performance. The Grad-CAM demonstrated that the DCNN placed focus on the necrotic region.

CONCLUSION

Compared with clinician-led diagnoses, the developed DCNN system is more accurate in diagnosing early ONFH, avoiding empirical dependence and inter-reader variability. Our findings support the integration of deep learning systems into real clinical settings to assist orthopaedic surgeons in diagnosing early ONFH.

Exploring the Risk Factors for the Misdiagnosis of Osteonecrosis of Femoral Head: A Case-Control Study

Objective

The purpose of the present study was to evaluate the present situation and risk factors for the misdiagnosis of osteonecrosis of femoral head (ONFH), providing the basis for accurate diagnosis of ONFH.

Methods

For this retrospective study, 1471 patients with ONFH were selected from the China Osteonecrosis of Femoral Head Database (CONFHD). These patients had been recruited between July 2016 and December 2018. According to whether or not they were misdiagnosed, the patients were divided into two groups, with 1168 cases (22–84 years old) included in the diagnosis group and 303 cases (21–80 years old) in the misdiagnosis group. Misdiagnosis was measured using the following criteria: (i) the patient had the same symptoms and signs, and the second diagnosis was not consistent with the initial diagnosis within 6 months; and (ii) the patient was admitted to a hospital participating in CONFHD and the previous diagnosis was inconsistent with the diagnosis given by the expert group. Comparisons of age, visual analogue scale for pain, and body mass index between the two groups were performed using a t‐test. Gender, causes of ONFH, primary diseases requiring corticosteroids, methods of corticosteroid use, corticosteroid species, type of trauma, onset side of the disease, pain side, whether symptoms are hidden, and type of imaging examination at the initial visit were compared using the χ²‐test. Years of alcohol consumption, weekly alcohol consumption, and physician title at the initial visit were compared using a Mann–Whitney U‐test. Furthermore, the statistically significant factors were evaluated using multiple regression analysis to investigate the risk factors of misdiagnosis.

Results

A total of 303 patients (20.6%) were misdiagnosed: 118 cases were misdiagnosed as lumbar disc herniation, 86 cases as hip synovitis, 48 cases as hip osteoarthritis, 32 cases as rheumatoid arthritis, 11 cases as piriformis syndrome, 5 cases as sciatica, and 3 cases as soft‐tissue injury. Whether symptoms are hidden (P = 0.038, odds ratio [OR] = 1.546, 95% confidence interval [CI] = 1.025–2.332), physician title at the initial visit (P < 0.001, OR = 3.324, 95% CI = 1.850–5.972), X‐ray examination (P < 0.001, OR = 4.742, 95% CI = 3.159–7.118), corticosteroids (P < 0.001, OR = 0.295, 95% CI = 0.163–0.534), alcohol (P < 0.001, OR = 0.305, 95% CI = 0.171–0.546), and magnetic resonance imaging (MRI) examination (P = 0.042, OR = 0.649, 95% CI = 0.427–0.985) were each found to be associated with misdiagnosis.

Conclusion

Osteonecrosis of the femoral head is easily misdiagnosed as lumbar disc herniation, hip synovitis, hip osteoarthritis, and rheumatoid arthritis. Patient history of corticosteroid use or alcohol abuse and MRI examination at the initial diagnosis may be protective factors for misdiagnosis. Hidden symptoms, physician title at the initial visit (as attending doctor or resident doctor), and only X‐ray examination at the initial diagnosis may be risk factors for misdiagnosis.