Machine learning-based computer-aided simple triage (CAST) for COVID-19 pneumonia as compared with triage by board-certified chest radiologists.

PURPOSE: Several reporting systems have been proposed for providing standardized language and diagnostic categories aiming for expressing the likelihood that lung abnormalities on CT images represent COVID-19. We developed a machine learning (ML)-based CT texture analysis software for simple triage based on the RSNA Expert Consensus Statement system. The purpose of this study was to conduct a multi-center and multi-reader study to determine the capability of ML-based computer-aided simple triage (CAST) software based on RSNA expert consensus statements for diagnosis of COVID-19 pneumonia. METHODS: For this multi-center study, 174 cases who had undergone CT and polymerase chain reaction (PCR) tests for COVID-19 were retrospectively included. Their CT data were then assessed by CAST and consensus from three board-certified chest radiologists, after which all cases were classified as either positive or negative. Diagnostic performance was then compared by McNemar's test. To determine radiological finding evaluation capability of CAST, three other board-certified chest radiologists assessed CAST results for radiological findings into five criteria. Finally, accuracies of all radiological evaluations were compared by McNemar's test. RESULTS: A comparison of diagnosis for COVID-19 pneumonia based on RT-PCR results for cases with COVID-19 pneumonia findings on CT showed no significant difference of diagnostic performance between ML-based CAST software and consensus evaluation (p > 0.05). Comparison of agreement on accuracy for all radiological finding evaluations showed that emphysema evaluation accuracy for investigator A (AC = 91.7%) was significantly lower than that for investigators B (100%, p = 0.0009) and C (100%, p = 0.0009). CONCLUSION: This multi-center study shows COVID-19 pneumonia triage by CAST can be considered at least as valid as that by chest expert radiologists and may be capable for playing as useful a complementary role for management of suspected COVID-19 pneumonia patients as well as the RT-PCR test in routine clinical practice.

Quantitative Computed Tomography Analysis of the Longitudinal Change Between Centrilobular and Paraseptal Emphysema Subtypes: A Retrospective Study.

OBJECTIVE: This study aimed to investigate the difference between the extent of centrilobular emphysema (CLE) and paraseptal emphysema (PSE) on follow-up chest CT scans and their relationship to the cross-sectional area (CSA) of small pulmonary vessels. METHODS: Sixty-two patients (36 CLE and 26 PSE) who underwent 2 chest CT scans were enrolled in this study. The percentage of low attenuation volume (%LAV) and total CSA of the small pulmonary vessels <5 mm 2 (%CSA < 5) were measured at the 2 time points. Analysis of the initial %CSA < 5 and the change in the %LAV and %CSA < 5 on follow-up imaging was performed. RESULTS: The initial %CSA < 5 was not significantly different between the CLE and the PSE groups (CLE, 0.66 vs. PSE, 0.71; P = 0.78). There was no significant difference in the longitudinal change in the %LAV between the 2 groups (CLE, -0.048% vs. PSE, 0.005%; P = 0.26). The longitudinal change in the %CSA < 5 in patients with PSE significantly decreased compared with those with CLE (CLE, 0.025% vs. PSE, -0.018%; P = 0.02). CONCLUSIONS: The longitudinal change in the %CSA < 5 was significantly different for patients with CLE and PSE, demonstrating an important pathophysiological difference between the subtypes.

Use of machine learning with two-dimensional synthetic mammography for axillary lymph node metastasis prediction in breast cancer: a preliminary study.

Background: As of 2020, breast cancer is the most common type of cancer and the fifth most common cause of cancer-related deaths worldwide. The non-invasive prediction of axillary lymph node (ALN) metastasis using two-dimensional synthetic mammography (SM) generated from digital breast tomosynthesis (DBT) could help mitigate complications related to sentinel lymph node biopsy or dissection. Thus, this study aimed to investigate the possibility of predicting ALN metastasis using radiomic analysis of SM images. Methods: Seventy-seven patients diagnosed with breast cancer using full-field digital mammography (FFDM) and DBT were included in the study. Radiomic features were calculated using segmented mass lesions. The ALN prediction models were constructed based on a logistic regression model. Parameters such as the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Results: The FFDM model yielded an AUC value of 0.738 [95% confidence interval (CI): 0.608-0.867], with sensitivity, specificity, PPV, and NPV of 0.826, 0.630, 0.488, and 0.894, respectively. The SM model yielded an AUC value of 0.742 (95% CI: 0.613-0.871), with sensitivity, specificity, PPV, and NPV of 0.783, 0.630, 0.474, and 0.871, respectively. No significant differences were observed between the two models. Conclusions: The ALN prediction model using radiomic features extracted from SM images demonstrated the possibility of enhancing the accuracy of diagnostic imaging when utilised together with traditional imaging techniques.

Radiomics model of diffusion-weighted whole-body imaging with background signal suppression (DWIBS) for predicting axillary lymph node status in breast cancer.

BACKGROUND: In breast cancer diagnosis and treatment, non-invasive prediction of axillary lymph node (ALN) metastasis can help avoid complications related to sentinel lymph node biopsy. OBJECTIVE: This study aims to develop and evaluate machine learning models using radiomics features extracted from diffusion-weighted whole-body imaging with background signal suppression (DWIBS) examination for predicting the ALN status. METHODS: A total of 100 patients with histologically proven, invasive, clinically N0 breast cancer who underwent DWIBS examination consisting of short tau inversion recovery (STIR) and DWIBS sequences before surgery were enrolled. Radiomic features were calculated using segmented primary lesions in DWIBS and STIR sequences and were divided into training (n = 75) and test (n = 25) datasets based on the examination date. Using the training dataset, optimal feature selection was performed using the least absolute shrinkage and selection operator algorithm, and the logistic regression model and support vector machine (SVM) classifier model were constructed with DWIBS, STIR, or a combination of DWIBS and STIR sequences to predict ALN status. Receiver operating characteristic curves were used to assess the prediction performance of radiomics models. RESULTS: For the test dataset, the logistic regression model using DWIBS, STIR, and a combination of both sequences yielded an area under the curve (AUC) of 0.765 (95% confidence interval: 0.548-0.982), 0.801 (0.597-1.000), and 0.779 (0.567-0.992), respectively, whereas the SVM classifier model using DWIBS, STIR, and a combination of both sequences yielded an AUC of 0.765 (0.548-0.982), 0.757 (0.538-0.977), and 0.779 (0.567-0.992), respectively. CONCLUSIONS: Use of machine learning models incorporating with the quantitative radiomic features derived from the DWIBS and STIR sequences can potentially predict ALN status.

The Utility of Automated ASPECTS in Acute Ischemic Stroke for Intravenous Recombinant Tissue Plasminogen Activator (IV-rtPA) Therapy.

PURPOSE: This study aimed to investigate the accuracy and clinical significance of an artificial intelligence (AI)-based automated Alberta Stroke Program Early Computed Tomography (ASPECT) scoring software of head CT for the indication of intravenous recombinant tissue plasminogen activator (rt-PA) therapy. METHODS: This study included two populations of acute ischemic stroke: one comprised patients who had undergone head CT within 48 h of presentation (Population #1, n = 448), while the other included patients within 4.5 h from onset (Population #2, n = 132). The primary endpoint was the concordance rate of ASPECTS of the neurologists and AI software against the benchmark score. The secondary endpoints were to validate the accuracy of the neurologist and AI software in assessing the ability to rule out extensive infarction (ASPECTS of 0-5) in population #2. RESULTS: The reading accuracy of AI software was comparable to that of the board-certified vascular neurologists. The detection rate of cardiogenic cerebral embolism was better than that of atherothrombotic cerebral infarction. By excluding extensive infarction, AI-software showed a higher specificity and equivalent sensitivity compared to those of experts. CONCLUSIONS: The AI software for ASPECTS showed convincing agreement with expert evaluation and would be supportive in determining the indications of intravenous rt-PA therapy.

Classifying presence or absence of calcifications on mammography using generative contribution mapping.

The purpose of this study was to verify the efficacy of generative contribution mapping (GCM), an explainable deep learning model for images, in classifying the presence or absence of calcifications on mammography. The learning dataset consisted of 303 full-field digital mammography (FFDM) images labeled with microcalcifications obtained from the public INbreast database without extremely dense images. FFDM images were divided into calcification and non-calcification patch images using a sliding window method with 25% overlap. The patch images of the mediolateral oblique (MLO) and craniocaudal (CC) views were divided into a training set of 70%, a validation set of 10%, and a testing set of 20%. The classification performance of GCM classifiers was evaluated and compared with that of EfficientNet classifiers. Visualization maps of GCM highlighted regions of interest more clearly than EfficientNet's gradient-weighted class activation maps. The results showed that GCM classifiers yielded an accuracy of 0.92 (CC), 0.91 (MLO), and an area under the receiver operating characteristic curve of 0.92 (CC), 0.94 (MLO). In conclusion, GCM could accurately classify the presence or absence of calcifications on mammograms and explain intuitively reasonable grounds for their classification with visualization maps highlighting regions of interest.

Development and external validation of a deep learning-based computed tomography classification system for COVID-19.

BACKGROUND: We aimed to develop and externally validate a novel machine learning model that can classify CT image findings as positive or negative for SARS-CoV-2 reverse transcription polymerase chain reaction (RT-PCR). METHODS: We used 2,928 images from a wide variety of case-control type data sources for the development and internal validation of the machine learning model. A total of 633 COVID-19 cases and 2,295 non-COVID-19 cases were included in the study. We randomly divided cases into training and tuning sets at a ratio of 8:2. For external validation, we used 893 images from 740 consecutive patients at 11 acute care hospitals suspected of having COVID-19 at the time of diagnosis. The dataset included 343 COVID-19 patients. The reference standard was RT-PCR. RESULTS: In external validation, the sensitivity and specificity of the model were 0.869 and 0.432, at the low-level cutoff, 0.724 and 0.721, at the high-level cutoff. Area under the receiver operating characteristic was 0.76. CONCLUSIONS: Our machine learning model exhibited a high sensitivity in external validation datasets and may assist physicians to rule out COVID-19 diagnosis in a timely manner at emergency departments. Further studies are warranted to improve model specificity.

Nonselective Bilateral Embolization of Internal Iliac Arteries with N-Butyl-2-Cyanoacrylate in Hemodynamically Unstable Patients with Pelvic Fracture.

Purpose: This study was designed to evaluate the efficacy and safety of nonselective bilateral embolization of the internal iliac arteries (IIAs) with n-butyl-2-cyanoacrylate (NBCA) in hemodynamically unstable patients with pelvic fractures. Material and Methods: Twelve patients underwent nonselective bilateral embolization of the IIAs using NBCA diluted with lipiodol at our institution between January 2004 and March 2014. We analyzed the time of bilateral occlusion of the IIAs, the time from admission to entrance into the interventional radiology room, the need for repeat embolization, outcomes, cause of death, follow-up period, and complications. Results: The mean duration of bilateral occlusion of the IIAs was 17 min (range, 4-34 min), and the mean time from admission to entrance into the interventional radiology room was 89 min (range, 28-168 min). All patients underwent technically successful embolization. Repeat embolization was required after treatment in three patients. The mortality rate was 33.3%. Complications after embolization were suspected in one patient. Conclusions: Nonselective bilateral embolization of IIAs with NBCA could be a choice of treatment for hemodynamically unstable patients with severe pelvic fracture hemorrhage.