Nonradiology Health-Care Professionals Significantly Benefit From AI Assistance in Emergency-Related Chest Radiography Interpretation.

BACKGROUND: Chest radiographs (CXRs) are still of crucial importance in primary diagnostics, but their interpretation poses difficulties at times. RESEARCH QUESTION: Can a convolutional neural network-based artificial intelligence (AI) system that interprets CXRs add value in an emergency unit setting? STUDY DESIGN AND METHODS: A total of 563 CXRs acquired in the emergency unit of a major university hospital were retrospectively assessed twice by three board-certified radiologists, three radiology residents, and three emergency unit-experienced nonradiology residents (NRRs). They used a two-step reading process: (1) without AI support (woAI); and (2) with AI support (wAI) providing additional images with AI overlays. Suspicion of four suspected pathologies (pleural effusion, pneumothorax, consolidations suspicious for pneumonia, and nodules) was reported on a five-point confidence scale. Confidence scores of the board-certified radiologists were converted into four binary reference standards (RFS I-IV) of different sensitivities. Performance by radiology residents and NRRs woAI/wAI were statistically compared by using receiver-operating characteristics (ROCs), Youden statistics, and operating point metrics derived from fitted ROC curves. RESULTS: NRRs could significantly improve performance, sensitivity, and accuracy wAI in all four pathologies tested. In the most sensitive RFS IV, NRR consensus improved the area under the ROC curve (mean, 95% CI) in the detection of the time-critical pathology pneumothorax from 0.846 (0.785-0.907) woAI to 0.974 (0.947-1.000) wAI (P < .001), which represented a gain of 30% in sensitivity and 2% in accuracy (while maintaining an optimized specificity). The most pronounced effect was observed in nodule detection, with NRR wAI improving sensitivity by 53% and accuracy by 7% (area under the ROC curve woAI, 0.723 [0.661-0.785]; wAI, 0.890 [0.848-0.931]; P < .001). The RR consensus wAI showed smaller, mostly nonsignificant gains in performance, sensitivity, and accuracy. INTERPRETATION: In an emergency unit setting without 24/7 radiology coverage, the presented AI solution features an excellent clinical support tool to nonradiologists, similar to a second reader, and allows for a more accurate primary diagnosis and thus earlier therapy initiation.

Automated MRI Lung Segmentation and 3D Morphologic Features for Quantification of Neonatal Lung Disease.

Purpose: To analyze the performance of deep learning (DL) models for segmentation of the neonatal lung in MRI and investigate the use of automated MRI-based features for assessment of neonatal lung disease. Materials and Methods: Quiet-breathing MRI was prospectively performed in two independent cohorts of preterm infants (median gestational age, 26.57 weeks; IQR, 25.3-28.6 weeks; 55 female and 48 male infants) with (n = 86) and without (n = 21) chronic lung disease (bronchopulmonary dysplasia [BPD]). Convolutional neural networks were developed for lung segmentation, and a three-dimensional reconstruction was used to calculate MRI features for lung volume, shape, pixel intensity, and surface. These features were explored as indicators of BPD and disease-associated lung structural remodeling through correlation with lung injury scores and multinomial models for BPD severity stratification. Results: The lung segmentation model reached a volumetric Dice coefficient of 0.908 in cross-validation and 0.880 on the independent test dataset, matching expert-level performance across disease grades. MRI lung features demonstrated significant correlations with lung injury scores and added structural information for the separation of neonates with BPD (BPD vs no BPD: average area under the receiver operating characteristic curve [AUC], 0.92 ± 0.02 [SD]; no or mild BPD vs moderate or severe BPD: average AUC, 0.84 ± 0.03). Conclusion: This study demonstrated high performance of DL models for MRI neonatal lung segmentation and showed the potential of automated MRI features for diagnostic assessment of neonatal lung disease while avoiding radiation exposure.Keywords: Bronchopulmonary Dysplasia, Chronic Lung Disease, Preterm Infant, Lung Segmentation, Lung MRI, BPD Severity Assessment, Deep Learning, Lung Imaging Biomarkers, Lung Topology Supplemental material is available for this article. Published under a CC BY 4.0 license.See also the commentary by Parraga and Sharma in this issue.

Immuno-histochemical correlation of fibrosis-related markers with the desmoplastic reaction of the mesentery in small intestine neuroendocrine neoplasms.

INTRODUCTION: Small intestine neuroendocrine neoplasms (siNENs) will attain more importance due to their increasing incidence. Moreover, siNENs might lead to a desmoplastic reaction (DR) of the mesentery causing severe complications and deteriorating prognosis. The expression of fibrosis-related proteins appears to be the key mechanisms for the development of this desmoplastic reaction. Therefore, this study aimed to investigate the association of the desmoplastic mesentery with specific fibrosis-related protein expression levels. MATERIALS AND METHODS: By immunohistochemistry, the protein expression levels of four fibrosis-related markers (APLP2, BNIP3L, CD59, DKK3) were investigated in primary tumors of 128 siNENs. The expression levels were correlated with the presence of a desmoplastic reaction and clinico-pathological parameters. RESULTS: In the primary tumor, APLP2, BNIP3L, CD59 and DKK3 were highly expressed in 29.7% (n = 38), 64.9% (n = 83), 92.2% (n = 118) and 80.5% (n = 103), respectively. There was no significant correlation of a single marker or the complete marker panel to the manifestation of a desmoplastic mesentery. The desmoplastic mesentery was significantly associated with clinical symptoms, such as flushing and diarrhea. However, neither the fibrosis-related marker panel nor single marker expressions were associated with clinical symptoms. DISCUSSION: The expression rates of four fibrosis-related markers in the primary tumor display a distinct pattern. However, the expression patterns are not associated with desmoplastic altered mesenteric lymph node metastases and the expression patterns did not correlate with prognosis. These findings suggest alternative mechanisms being responsible for the desmoplastic reaction.

Pelvic inlet area is associated with birth mode.

INTRODUCTION: To determine whether a pelvis is wide enough for spontaneous delivery has long been the subject of obstetric research. A number of variables have been proposed as predictors, all with limited accuracy. In this study, we use a novel three-dimensional (3D) method to measure the female pelvis and assess which pelvic features influence birth mode. We compare the 3D pelvic morphology of women who delivered vaginally, women who had cesarean sections, and nulliparous women. The aim of this study is to identify differences in pelvic morphology between these groups. MATERIAL AND METHODS: This observational study included women aged 50 years and older who underwent a CT scan of the pelvis for any medical indication. We recorded biometric data including height, weight, and age, and obtained the obstetric history. The bony pelvis was extracted from the CT scans and reconstructed in three dimensions. By placing 274 landmarks on each surface model, the pelvises were measured in detail. The pelvic inlet was measured using 32 landmarks. The trial was registered at the German Clinical Trials Register DRKS (DRKS00017690). RESULTS: For this study, 206 women were screened. Exclusion criteria were foreign material in the bony pelvis, unknown birth mode, and exclusively preterm births. Women who had both a vaginal birth and a cesarean section were excluded from the group comparison. We compared the pelvises of 177 women between three groups divided by obstetric history: vaginal births only (n = 118), cesarean sections only (n = 21), and nulliparous women (n = 38). The inlet area was significantly smaller in the cesarean section group (mean = 126.3 cm2 ) compared with the vaginal birth group (mean = 134.9 cm2 , p = 0.002). The nulliparous women were used as a control group: there was no statistically significant difference in pelvic inlet area between the nulliparous and vaginal birth groups. CONCLUSIONS: By placing 274 landmarks on a pelvis reconstructed in 3D, a very precise measurement of the morphology of the pelvis is possible. We identified a significant difference in pelvic inlet area between women with vaginal delivery and those with cesarean section. A unique feature of this study is the method of measurement of the bony pelvis that goes beyond linear distance measurements as used in previous pelvimetric studies.

Clinically focused multi-cohort benchmarking as a tool for external validation of artificial intelligence algorithm performance in basic chest radiography analysis.

Artificial intelligence (AI) algorithms evaluating [supine] chest radiographs ([S]CXRs) have remarkably increased in number recently. Since training and validation are often performed on subsets of the same overall dataset, external validation is mandatory to reproduce results and reveal potential training errors. We applied a multicohort benchmarking to the publicly accessible (S)CXR analyzing AI algorithm CheXNet, comprising three clinically relevant study cohorts which differ in patient positioning ([S]CXRs), the applied reference standards (CT-/[S]CXR-based) and the possibility to also compare algorithm classification with different medical experts' reading performance. The study cohorts include [1] a cohort, characterized by 563 CXRs acquired in the emergency unit that were evaluated by 9 readers (radiologists and non-radiologists) in terms of 4 common pathologies, [2] a collection of 6,248 SCXRs annotated by radiologists in terms of pneumothorax presence, its size and presence of inserted thoracic tube material which allowed for subgroup and confounding bias analysis and [3] a cohort consisting of 166 patients with SCXRs that were evaluated by radiologists for underlying causes of basal lung opacities, all of those cases having been correlated to a timely acquired computed tomography scan (SCXR and CT within < 90 min). CheXNet non-significantly exceeded the radiology resident (RR) consensus in the detection of suspicious lung nodules (cohort [1], AUC AI/RR: 0.851/0.839, p = 0.793) and the radiological readers in the detection of basal pneumonia (cohort [3], AUC AI/reader consensus: 0.825/0.782, p = 0.390) and basal pleural effusion (cohort [3], AUC AI/reader consensus: 0.762/0.710, p = 0.336) in SCXR, partly with AUC values higher than originally published ("Nodule": 0.780, "Infiltration": 0.735, "Effusion": 0.864). The classifier "Infiltration" turned out to be very dependent on patient positioning (best in CXR, worst in SCXR). The pneumothorax SCXR cohort [2] revealed poor algorithm performance in CXRs without inserted thoracic material and in the detection of small pneumothoraces, which can be explained by a known systematic confounding error in the algorithm training process. The benefit of clinically relevant external validation is demonstrated by the differences in algorithm performance as compared to the original publication. Our multi-cohort benchmarking finally enables the consideration of confounders, different reference standards and patient positioning as well as the AI performance comparison with differentially qualified medical readers.

The role of vagal innervation on the early development of postoperative ileus in mice.

BACKGROUND: Postoperative ileus (POI) involves an intestinal inflammatory response that is modulated by afferent and efferent vagal activation. We aimed to identify the potential influence of the vagus nerve on POI by tracking central vagal activation and its role for peripheral inflammatory changes during the early hours after surgery. METHODS: C57BL6 mice were vagotomized (V) 3-4 days prior to experiments, while control animals received sham vagotomy (SV). Subgroups underwent either laparotomy (sham operation; S-POI) or laparotomy followed by standardized small bowel manipulation to induce postoperative ileus (POI). Three hours and 9 h later, respectively, a jejunal segment was harvested and infiltration of inflammatory cells in intestinal muscularis was evaluated by fluorescein isothiocyanate (FITC) avidin and myeloperoxidase (MPO) staining. Moreover, the brain stem was harvested, and central nervous activation was investigated by Fos immunochemistry in both the nucleus of the solitary tract (NTS) and the area postrema (AP). Data are presented as mean ± SEM, and a p < 0.05 was considered statistically significant. KEY RESULTS: Three hour experiments revealed no significant differences between all experimental groups, except MPO staining: 3 h after abdominal surgery, there were significantly more MPO-positive cells in vagotomized S-POI animals compared to sham-vagotomized S-POI animals (26.7 ± 7.1 vs. 5.1 ± 2.4, p < 0.01). Nine hour postoperatively intramuscular mast cells (IMMC) were significantly decreased in the intestinal muscularis of V/POI animals compared to SV/POI animals (1.5 ± 0.3 vs. 5.9 ± 0.2, p < 0.05), while MPO-positive cells were increased in V/POI animals compared to SV/POI animals (713.2 ± 99.4 vs. 46.9 ± 5.8, p < 0.05). There were less Fos-positive cells in the NTS of V/POI animals compared to SV/POI animals (64.7 ± 7.8 vs. 132.8 ± 23.9, p < 0.05) and more Fos-positive cells in the AP of V/POI animals compared to SV/POI animals 9 h postoperatively (38.0 ± 2.0 vs. 13.7 ± 0.9, p < 0.001). CONCLUSIONS AND INTERFERENCES: Afferent nerve signaling to the central nervous system during the development of early POI seems to be mediated mainly via the vagus nerve and to a lesser degree via systemic circulation. During the early hours of POI, the intestinal immune response may be attenuated by vagal modulation, suggesting interactions between the central nervous system and the intestine.

Artificial Intelligence in Chest Radiography Reporting Accuracy: Added Clinical Value in the Emergency Unit Setting Without 24/7 Radiology Coverage.

OBJECTIVES: Chest radiographs (CXRs) are commonly performed in emergency units (EUs), but the interpretation requires radiology experience. We developed an artificial intelligence (AI) system (precommercial) that aims to mimic board-certified radiologists' (BCRs') performance and can therefore support non-radiology residents (NRRs) in clinical settings lacking 24/7 radiology coverage. We validated by quantifying the clinical value of our AI system for radiology residents (RRs) and EU-experienced NRRs in a clinically representative EU setting. MATERIALS AND METHODS: A total of 563 EU CXRs were retrospectively assessed by 3 BCRs, 3 RRs, and 3 EU-experienced NRRs. Suspected pathologies (pleural effusion, pneumothorax, consolidations suspicious for pneumonia, lung lesions) were reported on a 5-step confidence scale (sum of 20,268 reported pathology suspicions [563 images × 9 readers × 4 pathologies]) separately by every involved reader. Board-certified radiologists' confidence scores were converted into 4 binary reference standards (RFSs) of different sensitivities. The RRs' and NRRs' performances were statistically compared with our AI system (trained on nonpublic data from different clinical sites) based on receiver operating characteristics (ROCs) and operating point metrics approximated to the maximum sum of sensitivity and specificity (Youden statistics). RESULTS: The NRRs lose diagnostic accuracy to RRs with increasingly sensitive BCRs' RFSs for all considered pathologies. Based on our external validation data set, the AI system/NRRs' consensus mimicked the most sensitive BCRs' RFSs with areas under ROC of 0.940/0.837 (pneumothorax), 0.953/0.823 (pleural effusion), and 0.883/0.747 (lung lesions), which were comparable to experienced RRs and significantly overcomes EU-experienced NRRs' diagnostic performance. For consolidation detection, the AI system performed on the NRRs' consensus level (and overcomes each individual NRR) with an area under ROC of 0.847 referenced to the BCRs' most sensitive RFS. CONCLUSIONS: Our AI system matched RRs' performance, meanwhile significantly outperformed NRRs' diagnostic accuracy for most of considered CXR pathologies (pneumothorax, pleural effusion, and lung lesions) and therefore might serve as clinical decision support for NRRs.

Interpretation of Thoracic Radiography Shows Large Discrepancies Depending on the Qualification of the Physician-Quantitative Evaluation of Interobserver Agreement in a Representative Emergency Department Scenario.

(1) Background: Chest radiography (CXR) is still a key diagnostic component in the emergency department (ED). Correct interpretation is essential since some pathologies require urgent treatment. This study quantifies potential discrepancies in CXR analysis between radiologists and non-radiology physicians in training with ED experience. (2) Methods: Nine differently qualified physicians (three board-certified radiologists [BCR], three radiology residents [RR], and three non-radiology residents involved in ED [NRR]) evaluated a series of 563 posterior-anterior CXR images by quantifying suspicion for four relevant pathologies: pleural effusion, pneumothorax, pneumonia, and pulmonary nodules. Reading results were noted separately for each hemithorax on a Likert scale (0-4; 0: no suspicion of pathology, 4: safe existence of pathology) adding up to a total of 40,536 reported pathology suspicions. Interrater reliability/correlation and Kruskal-Wallis tests were performed for statistical analysis. (3) Results: While interrater reliability was good among radiologists, major discrepancies between radiologists' and non-radiologists' reading results could be observed in all pathologies. Highest overall interrater agreement was found for pneumothorax detection and lowest agreement in raising suspicion for malignancy suspicious nodules. Pleural effusion and pneumonia were often suspected with indifferent choices (1-3). In terms of pneumothorax detection, all readers mainly decided for a clear option (0 or 4). Interrater reliability was usually higher when evaluating the right hemithorax (all pathologies except pneumothorax). (4) Conclusions: Quantified CXR interrater reliability analysis displays a general uncertainty and strongly depends on medical training. NRR can benefit from radiology reporting in terms of time efficiency and diagnostic accuracy. CXR evaluation of long-time trained ED specialists has not been tested.

Is the EOS imaging system as accurate as conventional radiography in grading osteoarthritis of the knee?

OBJECTIVES: The EOS imaging system allows for the acquirement of long-leg radiographic images in a standing position without stitching artifacts or projection bias and at a comparatively low-radiation-dose exposure. The aim of our study was to compare the accuracy of EOS images of the lower limb to conventional radiographs (CR) of the knee in a.p. view for the grading of osteoarthritis (OA). METHODS: One hundred forty-two patients who had undergone EOS of the lower limb and radiography of the knee on the same day were included. For the grading of OA, the Kellgren and Lawrence score (KL) score and the Osteoarthritis Research Society International (OARSI) system were used. Additionally, the joint space was measured and compared between the two techniques. EOS images were compared to conventional anteroposterior radiographs of the knee which constitute the gold standard. RESULTS: Measurements of the joint space showed very good intra-class correlation. The calculated weighted kappa for the KL score of EOS versus CR was excellent. The comparison of the different parameters of the OARSI score showed superb weighted kappa scores between 0.9 and 0.96 (α < 0.001) for the parameters osteophytes and joint space narrowing. The parameter deformity showed a good agreement between EOS and radiographs (sensitivity 93.6%; specificity 100%). For the sclerosis parameter, an overall sensitivity of 71.3% and a specificity of 99.3% were calculated. CONCLUSIONS: The grading of OA using the KL score as well as the quantitative assessment of joint space width can be performed on EOS images in a.p. view as reliably as on CR. Subchondral sclerosis of the lateral and medial femur condyle or tibia is sometimes not as evident on EOS images. KEY POINTS: • Grading of OA may be performed as reliably with EOS images in a.p. view as with conventional radiographs in a.p. view. • EOS can be safely used for primary assessment of osteoarthritis of the knee. • In the preoperative setting for knee replacement surgery, conventional radiographs in two or three planes of the knee should still be acquired in addition to long-leg EOS images.