scholarly journals Deep Learning Could Diagnose Diabetic Nephropathy with Renal Pathological Immunofluorescent Images

Diagnostics ◽  
2020 ◽  
Vol 10 (7) ◽  
pp. 466
Author(s):  
Shinji Kitamura ◽  
Kensaku Takahashi ◽  
Yizhen Sang ◽  
Kazuhiko Fukushima ◽  
Kenji Tsuji ◽  
...  
Keyword(s):  
Diabetic Nephropathy ◽  
Deep Learning ◽  
Clinical Laboratory ◽  
Clinical Course ◽  
Laboratory Data ◽  
Area Under The Curve ◽  
Medical Doctors ◽  
Diagnostic Ratio ◽  
Interpretable Model ◽  
Microscopy Images

Artificial Intelligence (AI) imaging diagnosis is developing, making enormous steps forward in medical fields. Regarding diabetic nephropathy (DN), medical doctors diagnose them with clinical course, clinical laboratory data and renal pathology, mainly evaluate with light microscopy images rather than immunofluorescent images because there are no characteristic findings in immunofluorescent images for DN diagnosis. Here, we examined the possibility of whether AI could diagnose DN from immunofluorescent images. We collected renal immunofluorescent images from 885 renal biopsy patients in our hospital, and we created a dataset that contains six types of immunofluorescent images of IgG, IgA, IgM, C3, C1q and Fibrinogen for each patient. Using the dataset, 39 programs worked without errors (Area under the curve (AUC): 0.93). Five programs diagnosed DN completely with immunofluorescent images (AUC: 1.00). By analyzing with Local interpretable model-agnostic explanations (Lime), the AI focused on the peripheral lesion of DN glomeruli. On the other hand, the nephrologist diagnostic ratio (AUC: 0.75833) was slightly inferior to AI diagnosis. These findings suggest that DN could be diagnosed only by immunofluorescent images by deep learning. AI could diagnose DN and identify classified unknown parts with the immunofluorescent images that nephrologists usually do not use for DN diagnosis.

scholarly journals Differential Diagnostic Utility of Clinical Laboratory Data in Patients with Severe and Non-severe COVID-19

2021 ◽  
Vol In Press (In Press) ◽  
Author(s):  
Ali Hosseinzadeh ◽  
Mohammad Hassan Emamian ◽  
Marzieh Rohani-Rasaf ◽  
Ahmad Khosravi ◽  
Fariba Zare ◽  
...  
Keyword(s):  
Disease Severity ◽  
Clinical Laboratory ◽  
Characteristic Curve ◽  
Laboratory Data ◽  
Area Under The Curve ◽  
International Normalized Ratio ◽  
Therapeutic Effects ◽  
Neutrophil Lymphocyte Ratio ◽  
Laboratory Variables ◽  
Age Range

Background: Coronavirus disease 2019 (COVID-19) is a coronavirus outbreak caused by severe acute respiratory syndrome coronavirus 2 infection. Objectives: This study aimed to investigate the relationship between laboratory variables and COVID-19 severity. Methods: A total of 731 confirmed cases were included in this study. Based on the clinical course of the disease, the patients were divided into non-severe (n = 599) and severe (n = 132) groups. The area under the curve was estimated for each of the significant predictive factors by the receiver operating characteristic curve. Youden’s index was used to determine the optimal cut-off points to predict the severity of COVID-19. Results: Out of 731 patients, 407 (55.56%) cases were male. The mean age value and age range of the patients were 58.37 and 1 - 98 years, respectively. The age (OR = 1.03, 95% CI: 1.02 - 1.05), international normalized ratio (INR) (OR = 2.09, 95% CI: 1.11 - 3.96), lactate dehydrogenase (LDH) (OR = 1.003, 95% CI: 1.001 - 1.1.003), and neutrophil/lymphocyte ratio (NLR) (OR = 1.08, 95% CI: 1.02 - 1.14) were associated with the severity of COVID-19 in the multivariate analyses. The areas under the curve of LDH, NLR, and INR for the diagnosis of disease severity were 0.76, 0.69, and 0.62, respectively. Conclusions: The results of this study revealed that LDH, NLR, and INR could help to discriminate between non-severe and severe COVID-19 cases. Therefore, clinicians can use these variables to improve therapeutic effects and reduce disease severity.

A STUDY OF HEPATITIS B VIRUS GENOTYPES IN CHRONIC HEPATITIS B PATIENTS

2011 ◽  
pp. 25-29
Author(s):  
Keyword(s):  
Chronic Hepatitis ◽  
Hepatitis B ◽  
Chronic Hepatitis B ◽  
Clinical Laboratory ◽  
Laboratory Data ◽  
Hbv Dna ◽  
Hbv Genotypes ◽  
B Virus ◽  
Virus Genotypes ◽  
Genotype C

Aims: To measure the prevalence of HBV genotypes in chronic hepatitis B patients and their relation to HBeAg and HBV DNA level. Methods: 81 patients were enrolled in this study from January 2009 to December 2010. Clinical, laboratory data were collected during the patient’s hospitalization. Sera were quantitatively tested for HBeAg and HBV DNA. HBV genotyping was made by real-time PCR. Results: Among the 81 patients, 60.5% had genotype B, 26.7% had genotype C and 8.6% had mixed genotype B-C. Prevalence of symptoms (fatigue, anorexia, insomnia...) was higher in genotype C than in genotype B. Genotype C patients had positivity higher HBeAg than genotype B patients (56% vs. 38,8%, p <0.05). The rate of HBV DNA > 107 copies/mL was higher in genotype C group than in genotype B group (36% vs. 28,6%, p > 0.05). Conclusions: Most of the patients had genotypes B or C. Patients with genotype C had positive HBeAg and may be related to higher serological HBV DNA level than in genotype B.

scholarly journals Analyses of longitudinal, hospital clinical laboratory data with application to blood glucose concentrations

2011 ◽  
Vol 30 (27) ◽  
pp. 3208-3220 ◽  
Author(s):  
Jonathan S. Schildcrout ◽  
Sebastien Haneuse ◽  
Josh F. Peterson ◽  
Joshua C. Denny ◽  
Michael E. Matheny ◽  
...  
Keyword(s):  
Blood Glucose ◽  
Clinical Laboratory ◽  
Laboratory Data ◽  
Blood Glucose Concentrations

scholarly journals Explaining deep neural networks for knowledge discovery in electrocardiogram analysis

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Steven A. Hicks ◽  
Jonas L. Isaksen ◽  
Vajira Thambawita ◽  
Jonas Ghouse ◽  
Gustav Ahlberg ◽  
...  
Keyword(s):  
Decision Making ◽  
Deep Learning ◽  
Deep Neural Networks ◽  
Clinical Decision Making ◽  
Medical Knowledge ◽  
Clinical Decision ◽  
Medical Data ◽  
Medical Doctors ◽  
Medical Tests ◽  
Deep Learning Model

AbstractDeep learning-based tools may annotate and interpret medical data more quickly, consistently, and accurately than medical doctors. However, as medical doctors are ultimately responsible for clinical decision-making, any deep learning-based prediction should be accompanied by an explanation that a human can understand. We present an approach called electrocardiogram gradient class activation map (ECGradCAM), which is used to generate attention maps and explain the reasoning behind deep learning-based decision-making in ECG analysis. Attention maps may be used in the clinic to aid diagnosis, discover new medical knowledge, and identify novel features and characteristics of medical tests. In this paper, we showcase how ECGradCAM attention maps can unmask how a novel deep learning model measures both amplitudes and intervals in 12-lead electrocardiograms, and we show an example of how attention maps may be used to develop novel ECG features.

scholarly journals The association between clinical laboratory data and chest CT findings explains disease severity in a large Italian cohort of COVID-19 patients

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Simone Canovi ◽  
◽  
Giulia Besutti ◽  
Efrem Bonelli ◽  
Valentina Iotti ◽  
...  
Keyword(s):  
Clinical Laboratory ◽  
Laboratory Data ◽  
Organ Damage ◽  
Cross Sectional Study ◽  
Chest Ct ◽  
Arterial Oxygen ◽  
Emergency Rooms ◽  
Cross Sectional ◽  
Ct Findings ◽  
Patient Prognosis

Abstract Background Laboratory data and computed tomography (CT) have been used during the COVID-19 pandemic, mainly to determine patient prognosis and guide clinical management. The aim of this study was to evaluate the association between CT findings and laboratory data in a cohort of COVID-19 patients. Methods This was an observational cross-sectional study including consecutive patients presenting to the Reggio Emilia (Italy) province emergency rooms for suspected COVID-19 for one month during the outbreak peak, who underwent chest CT scan and laboratory testing at presentation and resulted positive for SARS-CoV-2. Results Included were 866 patients. Total leukocytes, neutrophils, C-reactive protein (CRP), creatinine, AST, ALT and LDH increase with worsening parenchymal involvement; an increase in platelets was appreciable with the highest burden of lung involvement. A decrease in lymphocyte counts paralleled worsening parenchymal extension, along with reduced arterial oxygen partial pressure and saturation. After correcting for parenchymal extension, ground-glass opacities were associated with reduced platelets and increased procalcitonin, consolidation with increased CRP and reduced oxygen saturation. Conclusions Pulmonary lesions induced by SARS-CoV-2 infection were associated with raised inflammatory response, impaired gas exchange and end-organ damage. These data suggest that lung lesions probably exert a central role in COVID-19 pathogenesis and clinical presentation.

scholarly journals Improved computer-aided detection of pulmonary nodules via deep learning in the sinogram domain

2019 ◽  
Vol 2 (1) ◽  
Author(s):  
Yongfeng Gao ◽  
Jiaxing Tan ◽  
Zhengrong Liang ◽  
Lihong Li ◽  
Yumei Huo
Keyword(s):  
Deep Learning ◽  
Area Under The Curve ◽  
Pulmonary Nodules ◽  
Ct Image ◽  
Computer Aided Detection ◽  
Ct Scanner ◽  
Human Visual Perception ◽  
Computer Aided ◽  
Detection And Diagnosis ◽  
Cade System

AbstractComputer aided detection (CADe) of pulmonary nodules plays an important role in assisting radiologists’ diagnosis and alleviating interpretation burden for lung cancer. Current CADe systems, aiming at simulating radiologists’ examination procedure, are built upon computer tomography (CT) images with feature extraction for detection and diagnosis. Human visual perception in CT image is reconstructed from sinogram, which is the original raw data acquired from CT scanner. In this work, different from the conventional image based CADe system, we propose a novel sinogram based CADe system in which the full projection information is used to explore additional effective features of nodules in the sinogram domain. Facing the challenges of limited research in this concept and unknown effective features in the sinogram domain, we design a new CADe system that utilizes the self-learning power of the convolutional neural network to learn and extract effective features from sinogram. The proposed system was validated on 208 patient cases from the publicly available online Lung Image Database Consortium database, with each case having at least one juxtapleural nodule annotation. Experimental results demonstrated that our proposed method obtained a value of 0.91 of the area under the curve (AUC) of receiver operating characteristic based on sinogram alone, comparing to 0.89 based on CT image alone. Moreover, a combination of sinogram and CT image could further improve the value of AUC to 0.92. This study indicates that pulmonary nodule detection in the sinogram domain is feasible with deep learning.

scholarly journals Defining Kawasaki disease and pediatric inflammatory multisystem syndrome-temporally associated to SARS-CoV-2 infection during SARS-CoV-2 epidemic in Italy: results from a national, multicenter survey

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Marco Cattalini ◽  
◽  
Sara Della Paolera ◽  
Fiammetta Zunica ◽  
Claudia Bracaglia ◽  
...  
Keyword(s):  
Kawasaki Disease ◽  
Clinical Laboratory ◽  
Troponin T ◽  
Inflammatory Diseases ◽  
Age At Onset ◽  
Laboratory Data ◽  
Clinical Manifestations ◽  
Chi Square ◽  
Pediatric Society ◽  
Multicenter Survey

Abstract Background There is mounting evidence on the existence of a Pediatric Inflammatory Multisystem Syndrome-temporally associated to SARS-CoV-2 infection (PIMS-TS), sharing similarities with Kawasaki Disease (KD). The main outcome of the study were to better characterize the clinical features and the treatment response of PIMS-TS and to explore its relationship with KD determining whether KD and PIMS are two distinct entities. Methods The Rheumatology Study Group of the Italian Pediatric Society launched a survey to enroll patients diagnosed with KD (Kawasaki Disease Group – KDG) or KD-like (Kawacovid Group - KCG) disease between February 1st 2020, and May 31st 2020. Demographic, clinical, laboratory data, treatment information, and patients’ outcome were collected in an online anonymized database (RedCAP®). Relationship between clinical presentation and SARS-CoV-2 infection was also taken into account. Moreover, clinical characteristics of KDG during SARS-CoV-2 epidemic (KDG-CoV2) were compared to Kawasaki Disease patients (KDG-Historical) seen in three different Italian tertiary pediatric hospitals (Institute for Maternal and Child Health, IRCCS “Burlo Garofolo”, Trieste; AOU Meyer, Florence; IRCCS Istituto Giannina Gaslini, Genoa) from January 1st 2000 to December 31st 2019. Chi square test or exact Fisher test and non-parametric Wilcoxon Mann-Whitney test were used to study differences between two groups. Results One-hundred-forty-nine cases were enrolled, (96 KDG and 53 KCG). KCG children were significantly older and presented more frequently from gastrointestinal and respiratory involvement. Cardiac involvement was more common in KCG, with 60,4% of patients with myocarditis. 37,8% of patients among KCG presented hypotension/non-cardiogenic shock. Coronary artery abnormalities (CAA) were more common in the KDG. The risk of ICU admission were higher in KCG. Lymphopenia, higher CRP levels, elevated ferritin and troponin-T characterized KCG. KDG received more frequently immunoglobulins (IVIG) and acetylsalicylic acid (ASA) (81,3% vs 66%; p = 0.04 and 71,9% vs 43,4%; p = 0.001 respectively) as KCG more often received glucocorticoids (56,6% vs 14,6%; p < 0.0001). SARS-CoV-2 assay more often resulted positive in KCG than in KDG (75,5% vs 20%; p < 0.0001). Short-term follow data showed minor complications. Comparing KDG with a KD-Historical Italian cohort (598 patients), no statistical difference was found in terms of clinical manifestations and laboratory data. Conclusion Our study suggests that SARS-CoV-2 infection might determine two distinct inflammatory diseases in children: KD and PIMS-TS. Older age at onset and clinical peculiarities like the occurrence of myocarditis characterize this multi-inflammatory syndrome. Our patients had an optimal response to treatments and a good outcome, with few complications and no deaths.

scholarly journals Machine learning applied on chest x-ray can aid in the diagnosis of COVID-19: a first experience from Lombardy, Italy

2021 ◽  
Vol 5 (1) ◽  
Author(s):  
Isabella Castiglioni ◽  
Davide Ippolito ◽  
Matteo Interlenghi ◽  
Caterina Beatrice Monti ◽  
Christian Salvatore ◽  
...  
Keyword(s):  
Deep Learning ◽  
Area Under The Curve ◽  
Clinical Information ◽  
Training Dataset ◽  
Independent Dataset ◽  
X Ray ◽  
Learning Classifier ◽  
Chest X Ray ◽  
Polymerase Chain ◽  
Deep Learning Model

Abstract Background We aimed to train and test a deep learning classifier to support the diagnosis of coronavirus disease 2019 (COVID-19) using chest x-ray (CXR) on a cohort of subjects from two hospitals in Lombardy, Italy. Methods We used for training and validation an ensemble of ten convolutional neural networks (CNNs) with mainly bedside CXRs of 250 COVID-19 and 250 non-COVID-19 subjects from two hospitals (Centres 1 and 2). We then tested such system on bedside CXRs of an independent group of 110 patients (74 COVID-19, 36 non-COVID-19) from one of the two hospitals. A retrospective reading was performed by two radiologists in the absence of any clinical information, with the aim to differentiate COVID-19 from non-COVID-19 patients. Real-time polymerase chain reaction served as the reference standard. Results At 10-fold cross-validation, our deep learning model classified COVID-19 and non-COVID-19 patients with 0.78 sensitivity (95% confidence interval [CI] 0.74–0.81), 0.82 specificity (95% CI 0.78–0.85), and 0.89 area under the curve (AUC) (95% CI 0.86–0.91). For the independent dataset, deep learning showed 0.80 sensitivity (95% CI 0.72–0.86) (59/74), 0.81 specificity (29/36) (95% CI 0.73–0.87), and 0.81 AUC (95% CI 0.73–0.87). Radiologists’ reading obtained 0.63 sensitivity (95% CI 0.52–0.74) and 0.78 specificity (95% CI 0.61–0.90) in Centre 1 and 0.64 sensitivity (95% CI 0.52–0.74) and 0.86 specificity (95% CI 0.71–0.95) in Centre 2. Conclusions This preliminary experience based on ten CNNs trained on a limited training dataset shows an interesting potential of deep learning for COVID-19 diagnosis. Such tool is in training with new CXRs to further increase its performance.

Sign in / Sign up

Export Citation Format

Share Document