Localizing thoracic diseases on chest X-ray plays a critical role in clinical practices such as diagnosis and treatment planning. However, current deep learning based approaches often require strong supervision, e.g. annotated bounding boxes, for training such systems, which is infeasible to harvest in large-scale. We present Probabilistic Class Activation Map (PCAM) pooling, a novel global pooling operation for lesion localization with only image-level supervision. PCAM pooling explicitly leverages the excellent localization ability of CAM [10] during training in a probabilistic fashion. Experiments on the ChestX-ray14 [7] dataset show a ResNet-34 [1] model trained with PCAM pooling outperforms state-of-the-art baselines on both the classification task and the localization task. Visual examination on the probability maps generated by PCAM pooling shows clear and sharp boundaries around lesion regions compared to the localization heatmaps generated by CAM. PCAM pooling is open sourced at https://github.com/jfhealthcare/Chexpert.
Objectives An artificial intelligence model was adopted to identify mild COVID-19 pneumonia from computed tomography (CT) volumes, and its diagnostic performance was then evaluated. Methods In this retrospective multicenter study, an atrous convolution-based deep learning model was established for the computer-assisted diagnosis of mild COVID-19 pneumonia. The dataset included 2087 chest CT exams collected from four hospitals between 1 January 2019 and 31 May 2020. The true positive rate, true negative rate, receiver operating characteristic curve, area under the curve (AUC) and convolutional feature map were used to evaluate the model. Results The proposed deep learning model was trained on 1538 patients and tested on an independent testing cohort of 549 patients. The overall sensitivity was 91.5% (195/213; p < 0.001, 95% CI: 89.2-93.9%), the overall specificity was 90.5% (304/ 336; p < 0.001, 95% CI: 88.0-92.9%) and the general AUC value was 0.955 (p < 0.001). Conclusions A deep learning model can accurately detect COVID-19 and serve as an important supplement to the COVID-19 reverse transcription-polymerase chain reaction (RT-PCR) test. Key Points • The implementation of a deep learning model to identify mild COVID-19 pneumonia was confirmed to be effective and feasible. • The strategy of using a binary code instead of the region of interest label to identify mild COVID-19 pneumonia was verified. • This AI model can assist in the early screening of COVID-19 without interfering with normal clinical examinations. Keywords Computer-assisted diagnosis. Volume CT. COVID-19. Artificial intelligence. Deep learning Abbreviations AUC Area under the curve CAD Computer-assisted diagnosis CAP Community-acquired pneumonia COVID-19 Coronavirus disease 2019 IgG Immunoglobulin G IgM Immunoglobulin M ROC Receiver operating characteristic RT-PCR Reverse transcription-polymerase chain reaction SSAC Sparse separable atrous convolution Jin-Cao Yao and Tao Wang contributed equally to this work.
Background Gastric adenocarcinoma of the fundic gland type (GA-FG) has been added to the 2019 edition of the World Health Organization’s list of digestive system-associated cancers. This lesion differentiates toward the fundic gland and mostly involves chief cell-predominant differentiation with low-grade cytology. Clinicians and pathologists are still unaware of this rare disease; consequently, some cases are incorrectly diagnosed. This study aimed to investigate the clinicopathological features of GA-FG using retrospective analyses of endoscopic and pathological findings. Materials and methods Samples were collected from patients diagnosed with GA-FG. The clinical courses of all patients were monitored prospectively and reviewed retrospectively. Available clinical information, endoscopic features, pathological appearance, and follow-up data were assessed. Immunohistochemistry [mucin (MUC) 2, MUC5, MUC6, P53, CDX2, Ki67, SYN, CD56, CGA, β-catenin, and pepsinogen-I] was examined using Envision two-step method.Results Eight cases of endoscopic submucosal dissection (ESD) were obtained from our institution. Patient age ranged from 48-80 years (mean, 65 years). Some patients were on acid-suppressing medication. Most lesions were located in the upper third (n = 7) and one was in the middle third of the stomach. Six lesions were of the superficial flat type, whereas two were of the superficial elevated type. Narrow-band imaging using magnifying endoscopy showed irregular microvascular patterns (MVPs) in four cases and regular MVPs in the remaining cases. All lesions were primarily solitary and ~6 mm in diameter (largest, 12 mm). The main body of the tumors were localized in the mucosal layer, of which six cases invade into the submucosal layer. Well-formed glands of chief cells were predominant. Tumor cells were positive for pepsinogen-I, MUC6, SYN, and CD56. Lymphatic and vascular infiltration and metastatic and recurrent disease were not observed in any case.Conclusion GA-FG, a well-differentiated adenocarcinoma with mild atypia, can be completely removed using ESD, with a favorable prognosis in patients.
Background: Primary gastrointestinal diffuse large B-cell lymphoma (PGI-DLBCL) lacks specific clinical manifestations and its malignancy renders prognostication and choice of treatment strategy difficult. The aim of this study was to evaluate microRNA (miR)-21 as potential non-invasive biomarkers for prognosis in PGI-DLBCL patients. Methods: Serum miR-21 expression in de novo PGI-DLBCL patients, consecutively enrolled for this study, was detected by quantitative real-time polymerase chain reaction (qRT-PCR). Relative expression was calculated using the comparative Ct method. Statistical significance was determined using the Mann-Whitney rank sum and Fisher’s exact test. Survival analysis was conducted using the Kaplan-Meier method. Results: Compared with healthy controls, serum miR-21 levels were significantly elevated in the PGI-DLBCL patients (n=156). The high expression level of serum miR-21 at diagnosis was associated with worse progression-free survival (PFS) (30 (9–42) vs 42 (12–52) months in high and low miR-21 groups) and overall survival (OS) (35 (15–52) vs 48 (17–61) months in high and low miR-21 groups) and was an independent risk factor for PFS and OS (hazard ratios 4.345 and 3.311, respectively). Furthermore, Bcl-2, Bcl-6 and Ki-67 were independently and positively associated with miR-21 expression. Conclusions: Our results suggest that miR-21 is a potential prognostic marker to predict clinical outcomes in PGI-DLBCL patients and a high miR-21 level is associated with poor outcomes.
A pilot-scale multi-tube dielectric barrier discharge (DBD) reactor coupled with a series of MnÀ Ag based catalysts for VOCs abatement was set up and investigated. The DBD with 2 %Ag-MnO x /Al 2 O 3 catalyst exhibited the best performance of removal efficiency (85.4 %) and energy yield (5.2 g/kWh) of toluene at the specific energy density of 39.7 J/L, much higher than the single DBD reactor (34.9 %, 2.1 g/kWh). Moreover, the degradation efficiencies of styrene, dimethyl, ethyl acetate and xylene in the plasma catalysis system were also studied. The catalyst presented high stability during the 50 hours of toluene plasma-catalysis oxidation. Finally, the physicochemical properties of this catalyst were characterized and discussed using an energy dispersive spectrometer, X-ray diffraction patterns, and X-ray photoelectron spectroscopy.
BackgroundGastric adenocarcinoma of the fundic gland type (GA-FG) has been added to the 2019 edition of the World Health Organization’s list of digestive system-associated cancers. This lesion differentiates toward the fundic gland and mostly involves chief cell-predominant differentiation with low-grade cytology. Clinicians and pathologists are still unaware of this rare disease; consequently, some cases are incorrectly diagnosed. This study aimed to investigate the clinicopathological features of GA-FG using retrospective analyses of endoscopic and pathological findings. Materials and methodsSamples were collected from patients diagnosed with GA-FG. The clinical courses of all patients were monitored prospectively and reviewed retrospectively. Available clinical information, endoscopic features, pathological appearance, and follow-up data were assessed. Immunohistochemistry [mucin (MUC) 2, MUC5, MUC6, P53, CDX2, Ki67, SYN, CD56, CGA, β-catenin, and pepsinogen-I] was examined using Envision two-step method.ResultsEight cases of endoscopic submucosal dissection (ESD) were obtained from our institution. Patient age ranged from 48-80 years (mean, 65 years). Some patients were on acid-suppressing medication. Most lesions were located in the upper third (n = 7) and one was in the middle third of the stomach. Six lesions were of the superficial flat type, whereas two were of the superficial elevated type. Narrow-band imaging using magnifying endoscopy showed irregular microvascular patterns (MVPs) in four cases and regular MVPs in the remaining cases. All lesions were primarily solitary and ~6 mm in diameter (largest, 12 mm). All tumors were localized in the mucosal layer with six cases of invasion into the submucosal layer. Well-formed glands of chief cells were predominant. Tumor cells were positive for pepsinogen-I, MUC6, SYN, and CD56. Lymphatic and vascular infiltration and metastatic and recurrent disease were not observed in any case.CONCLUSIONGA-FG, a well-differentiated adenocarcinoma with mild atypia, can be completely removed using ESD, with a favorable prognosis in patients.
Background: Lung endometriosis is an extremely rare gynecological disease. Literature reports suggest that most patients will show generic symptoms such as hemoptysis, pneumothorax, and hemopneumothorax, which can often result in misdiagnosis. There are case reports of 18 patients with lung endometriosis describing clinical manifestation, imaging changes, treatment and prognosis. To provide further information on this rare disease, we present a case of pulmonary endometriosis and a review of lung endometriosis.Case presentation: We report on a 19-year-old female who was admitted to hospital due to repeated menstrual hemoptysis for 3 months. Computed tomography during menstruation showed patchy high-density shadows sized approximately 0.5 cm × 0.5 cm × 0.5 cm in the right middle lobe of the lung. Following menstruation, hemoptysis and changes on CT imaging disappeared. Thoracoscopic right middle lobectomy, right lower lung repair, and closed thoracic drainage were performed. Postoperative histopathology confirmed lung endometriosis. There was no recurrence of symptoms after 6 months of follow-up. Conclusions: We review the literature on factors associated with lung endometriosis, diagnosis, and treatment options. We propose that the diagnosis for lung endometriosis should be made by comprehensively integrating patient reproductive history, clinical and imaging details as well as histopathology. Surgical resection appears to be an effective treatment for lung endometriosis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.