Elevating healthcare through artificial intelligence: analyzing the abdominal emergencies data set (TR_ABDOMEN_RAD_EMERGENCY) at TEKNOFEST-2022.

OBJECTIVES: The artificial intelligence competition in healthcare at TEKNOFEST-2022 provided a platform to address the complex multi-class classification challenge of abdominal emergencies using computer vision techniques. This manuscript aimed to comprehensively present the methodologies for data preparation, annotation procedures, and rigorous evaluation metrics. Moreover, it was conducted to introduce a meticulously curated abdominal emergencies data set to the researchers. METHODS: The data set underwent a comprehensive central screening procedure employing diverse algorithms extracted from the e-Nabiz (Pulse) and National Teleradiology System of the Republic of Türkiye, Ministry of Health. Full anonymization of the data set was conducted. Subsequently, the data set was annotated by a group of ten experienced radiologists. The evaluation process was executed by calculating F1 scores, which were derived from the intersection over union values between the predicted bounding boxes and the corresponding ground truth (GT) bounding boxes. The establishment of baseline performance metrics involved computing the average of the highest five F1 scores. RESULTS: Observations indicated a progressive decline in F1 scores as the threshold value increased. Furthermore, it could be deduced that class 6 (abdominal aortic aneurysm/dissection) was relatively straightforward to detect compared to other classes, with class 5 (acute diverticulitis) presenting the most formidable challenge. It is noteworthy, however, that if all achieved outcomes for all classes were considered with a threshold of 0.5, the data set's complexity and associated challenges became pronounced. CONCLUSION: This data set's significance lies in its pioneering provision of labels and GT-boxes for six classes, fostering opportunities for researchers. CLINICAL RELEVANCE STATEMENT: The prompt identification and timely intervention in cases of emergent medical conditions hold paramount significance. The handling of patients' care can be augmented, while the potential for errors is minimized, particularly amidst high caseload scenarios, through the application of AI. KEY POINTS: • The data set used in artificial intelligence competition in healthcare (TEKNOFEST-2022) provides a 6-class data set of abdominal CT images consisting of a great variety of abdominal emergencies. • This data set is compiled from the National Teleradiology System data repository of emergency radiology departments of 459 hospitals. • Radiological data on abdominal emergencies is scarce in literature and this annotated competition data set can be a valuable resource for further studies and new AI models.

Comparison of Different Slice Thicknesses on Chest Computed Tomography for Evaluation of COVID-19- Associated Pneumonia.

OBJECTIVE: To compare the value of chest computed tomography at 1-mm and 5-mm slice thickness in terms of computed tomography severity score and computed tomography evaluation time in the diagnosis of COVID-19. MATERIAL AND METHODS: Sixty-five patients were included in the study group who are reverse-transcription polymerase chain reaction-positive for COVID-19 and had chest computed tomography. The 1 mm and 5 mm reconstructed images were evaluated in 2 different sessions with 4-week intervals by 2 certificated general radiologists. The presence of COVID-19-related findings, COVID-19 final category, and evaluation time were recorded. Thin and thick slices were compared for these variables and inter-reader reliability calculated with the Statistical Package for the Social Sciences (SPSS) for Windows. RESULTS: There was no significant difference between the COVID-19-related findings on thorax computed tomography between 1-mm and 5-mm slices except crazy paving appearance, microvascular enlargement, and septal thickening. The frequency of the final categories of computed tomography results was consistent between the thick and thin slices. The computed tomography assessment time was significantly lower in 5 mm slices. The inter-reader reproducibility analysis results demonstrated good and excellent reproducibility of measurements between readers for both slice thicknesses. CONCLUSION: It was found that 5-mm reconstruction thickness of chest computed tomography can be employed for the initial detection of COVID-19-related findings and the final diagnostic category-related COVID-19 rather than 1-mm slices with a faster availability of results which can be beneficial on pandemic hospitals.

The utility of magnetic resonance angiography in children with nutcracker syndrome

Background/aim: The presented study aimed to evaluate the utility of magnetic resonance angiography (MRA) in the pediatric population with nutcracker syndrome. Materials and methods: Patients with suggestive clinical symptoms and laboratory findings and got the diagnosis of nutcracker syndrome with Doppler ultrasonography between January 2011­2019 were included in the study. In addition, children who had renal MRA due to hypertension were evaluated as the control group. MRA images of all patients were examined retrospectively by three radiologists at different levels of experience, and the superior mesenteric artery angle, aorta-mesenteric distance, left renal vein diameter both in the regions of aorta-mesenteric, and renal hilum were recorded. Results: Forty-five patients diagnosed with nutcracker syndrome were included in the study. The mean age of patients was 12 (4­16) and 30 (67%) were female. As the control group, 25 patients with hypertension who had MRA were included and they had a mean age of 12 (1­18) and 19 (76%) were male. The mean superior mesenteric artery angle was 26.5 ° (16­73 ± 12) in the patient group and 57.8 ° (25­139, ± 33) in the control group (p < 0.001); the mean aorta-mesenteric distance was 3.3 mm (1.7­6.5, ± 1.1) in the patient group and 8 mm (3.4­32, ± 5.9) in the control group (p < 0.001). MRA measurements of three radiologists were consistent with each other. Conclusion: MRA imaging can be applied as an alternative diagnostic method for Doppler ultrasonography and multidetector CT examinations by radiologists with different experience levels in pediatric patients with nutcracker syndrome.

Diagnostic efficacy of routine contrast-enhanced abdominal CT for the assessment of osteoporosis in the Turkish population

Background/aim: This study aimed to evaluate the diagnostic efficacy of vertebral Hounsfield unit (HU) values on routine contrast- enhanced abdominal computed tomography (CT) scans for the assessment of osteoporosis using dual-energy X-ray absorptiometry (DXA) T-scores as a reference standard. Materials and methods: A total of 111 consecutive patients who underwent contrast-enhanced abdominal CT examination for any indication and DXA within a 6-month period were retrospectively analyzed. The CT attenuation values of trabecular bone in HUs were measured in axial and sagittal planes from lumbar vertebrae 1­4 (L1­L4). The correlation between the DXA T-scores and HU values measured on the CT scans was evaluated by Pearson's correlation test. Areas under the curves (AUCs) were calculated by receiver operating characteristic analysis for diagnostic proficiency, and threshold values were determined. Paired t-test and Bland­Altman plot test were used to evaluate the correlation between axial and sagittal HU values. Results: There was a strong correlation between the DXA T-scores and HU values of all the lumbar vertebrae (P < 0.001). The highest correlation was for the L3 vertebra; L3 was thus chosen for additional analyses. The mean axial and sagittal L3 attenuations were 133.7 HU and 131.9 HU, respectively. The axial measurements were not significantly different from the sagittal measurements, with a mean difference of 1.8 HU (P > 0.05). The L3 axial CT attenuation threshold for 90% sensitivity was 170 HU and that for 90% specificity was 102 HU for distinguishing osteoporosis from osteopenia and normal bone mineral density (BMD). To distinguish the low BMD group from the normal group, the L3 axial CT attenuation threshold for approximately 90% sensitivity was 102 HU and for 90% specificity was 165 HU. Conclusion: The HUs derived from routine contrast-enhanced abdominal CT scans can be used for the evaluation of osteoporosis, without additional radiation exposure and cost.