Search

To describe and validate an artificial intelligence (AI)-driven structured reporting system by direct comparison of automatically generated reports to results from actual clinical reports generated by nuclear cardiology experts. Quantitative parameters extracted from myocardial perfusion imaging (MPI) studies are used by our AI reporting system to generate automatically a guideline-compliant structured report (sR). A new nonparametric approach generates distribution functions of rest and stress, perfusion, and thickening, for each of 17 left ventricle segments that are then transformed to certainty factors (CFs) that a segment is hypoperfused, ischemic. These CFs are then input to our set of heuristic rules used to reach diagnostic findings and impressions propagated into a sR referred as an AI-driven structured report (AIsR). The diagnostic accuracy of the AIsR for detecting coronary artery disease (CAD) and ischemia was tested in 1,000 patients who had undergone rest/stress SPECT MPI. At the high-specificity (SP) level, in a subset of 100 patients, there were no statistical differences in the agreements between the AIsr, and nine experts’ impressions of CAD (P = .33) or ischemia (P = .37). This high-SP level also yielded the highest accuracy across global and regional results in the 1,000 patients. These accuracies were statistically significantly better than the other two levels [sensitivity (SN)/SP tradeoff, high SN] across all comparisons. This AI reporting system automatically generates a structured natural language report with a diagnostic performance comparable to those of experts.

Congenital anomalies of the muscles of the lower extremity are rare. A case of complete absence of the semimembranosus muscle incidentally found with magnetic resonance imaging is reported. The patient was a 55-year-old female presenting with knee pain and no previous history of trauma. Clinical and imaging findings were consistent with meniscal tears. Two cases of uncommon distal insertion of the semimembranosus tendon are also described to illustrate the anatomical variations in this area. The anatomical and biomechanical relations between the semimembranosus muscle and the posteromedial corner of the knee are examined in this case report.

This study aims (1) to assess the prevalence and distribution of multiple occult injuries of the carpal bones and the distal forearm in patients with wrist pain and negative radiographs following trauma and (2) to evaluate the distribution and significance of joint effusions in the wrists with multiple osseous injuries. One hundred and thirty-one subjects, 74 men and 57 women, were consecutively examined in two institutions. All were acute trauma patients with negative X-rays whose clinical examination suggested possible fracture at the wrist or the distal forearm. Magnetic resonance (MR) wrist imaging was performed with and without fat saturation sequences. The MR images were analysed for detection of occult trabecular contusions and cortical discontinuity in the carpus, the distal forearm and the metacarpal bases. The prevalence and distribution of the injuries were assessed along with the distribution of joint effusions. Eight patients were excluded due to inadequate image quality. Two patients had bilateral injury. A total of 125 wrists were analysed. Seventy-eight (62.4%) wrists had occult bone injuries. Among these 78, 53 (68%) wrists had more than one injured bone. Twenty-five wrists (32%) had one injured bone. The highest number of injured bones per wrist was six. Injuries with a visible fracture line were seen in 29 (37.1%) wrists on MRI. The distal radius was the most frequent location for occult fracture line (11 cases). The injuries without a fracture line (contusion) were present in 49 (63%) wrists; they were detected more frequently in the scaphoid (35 cases). The lunate (29 cases) and the triquetrum (26 cases) were almost equally affected. The bone that was less frequently injured was the pisiform (four cases). Joint effusions were present in all 53 wrists with multiple bone lesions but more often situated in the ulnocarpal space [27 (50.3%) wrists]. There was no correlation between effusions in multiple locations (grades III and IV) and multiple bone injuries. This study revealed the presence of multifocal trabecular contusions without correlation with increased joint effusions in patients with negative radiographs and persistent pain. The clinical significance of these findings deserves further investigation.