RESUMEN
BACKGROUND: Research into artificial intelligence (AI)-based fracture detection in children is scarce and has disregarded the detection of indirect fracture signs and dislocations. OBJECTIVE: To assess the diagnostic accuracy of an existing AI-tool for the detection of fractures, indirect fracture signs, and dislocations. MATERIALS AND METHODS: An AI software, BoneView (Gleamer, Paris, France), was assessed for diagnostic accuracy of fracture detection using paediatric radiology consensus diagnoses as reference. Radiographs from a single emergency department were enrolled retrospectively going back from December 2021, limited to 1,000 radiographs per body part. Enrolment criteria were as follows: suspected fractures of the forearm, lower leg, or elbow; age 0-18 years; and radiographs in at least two projections. RESULTS: Lower leg radiographs showed 607 fractures. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were high (87.5%, 87.5%, 98.3%, 98.3%, respectively). Detection rate was low for toddler's fractures, trampoline fractures, and proximal tibial Salter-Harris-II fractures. Forearm radiographs showed 1,137 fractures. Sensitivity, specificity, PPV, and NPV were high (92.9%, 98.1%, 98.4%, 91.7%, respectively). Radial and ulnar bowing fractures were not reliably detected (one out of 11 radial bowing fractures and zero out of seven ulnar bowing fractures were correctly detected). Detection rate was low for styloid process avulsions, proximal radial buckle, and complete olecranon fractures. Elbow radiographs showed 517 fractures. Sensitivity and NPV were moderate (80.5%, 84.7%, respectively). Specificity and PPV were high (94.9%, 93.3%, respectively). For joint effusion, sensitivity, specificity, PPV, and NPV were moderate (85.1%, 85.7%, 89.5%, 80%, respectively). For elbow dislocations, sensitivity and PPV were low (65.8%, 50%, respectively). Specificity and NPV were high (97.7%, 98.8%, respectively). CONCLUSIONS: The diagnostic performance of BoneView is promising for forearm and lower leg fractures. However, improvement is mandatory before clinicians can rely solely on AI-based paediatric fracture detection using this software.
Asunto(s)
Luxaciones Articulares , Fracturas del Radio , Fracturas de Salter-Harris , Fracturas del Cúbito , Humanos , Niño , Recién Nacido , Lactante , Preescolar , Adolescente , Inteligencia Artificial , Estudios Retrospectivos , Fracturas del Radio/diagnóstico por imagen , Fracturas del Cúbito/diagnóstico por imagen , RadiografíaRESUMEN
BACKGROUND: Histopathological B3 lesions after minimal invasive breast biopsy (VABB) are a particular challenge for the clinician, as there are currently no binding recommendations regarding the subsequent procedure. PURPOSE: To analyze all B3 lesions, diagnosed at VABB and captured in the national central Swiss MIBB database and to provide a data basis for further management in this subgroup of patients. MATERIAL AND METHODS: All 9,153 stereotactically, sonographically, or magnetic resonance imaging (MRI)-guided vacuum-assisted breast biopsies, performed in Switzerland between 2009 and 2011, captured in a central database, were evaluated. The rate of B3 lesions and the definitive pathological findings in patients who underwent surgical resection were analyzed. RESULTS: The B3 rate was 17.0% (1532 of 9000 biopsies with B classification). Among the 521 lesions with a definitive postoperative diagnosis, the malignancy rate (invasive carcinoma or DCIS) was 21.5%. In patients with atypical ductal hyperplasia, papillary lesions, flat epithelial atypia, lobular neoplasia, and radial scar diagnosed by VABB, the malignancy rates were 25.9%, 3.1%, 18.3%, 26.4%, and 11.1%, respectively. CONCLUSION: B3 lesions, comprising 17%, of all analyzed biopsies, were common and the proportion of malignancies in those lesions undergoing subsequent surgical excision was high (21.5%).