ABSTRACT
Background: The previous COVID-19 lung diagnosis system lacks both scientific validation and the role of explainable artificial intelligence (AI) for understanding lesion localization. This study presents a cloud-based explainable AI, the "COVLIAS 2.0-cXAI" system using four kinds of class activation maps (CAM) models. Methodology: Our cohort consisted of ~6000 CT slices from two sources (Croatia, 80 COVID-19 patients and Italy, 15 control patients). COVLIAS 2.0-cXAI design consisted of three stages: (i) automated lung segmentation using hybrid deep learning ResNet-UNet model by automatic adjustment of Hounsfield units, hyperparameter optimization, and parallel and distributed training, (ii) classification using three kinds of DenseNet (DN) models (DN-121, DN-169, DN-201), and (iii) validation using four kinds of CAM visualization techniques: gradient-weighted class activation mapping (Grad-CAM), Grad-CAM++, score-weighted CAM (Score-CAM), and FasterScore-CAM. The COVLIAS 2.0-cXAI was validated by three trained senior radiologists for its stability and reliability. The Friedman test was also performed on the scores of the three radiologists. Results: The ResNet-UNet segmentation model resulted in dice similarity of 0.96, Jaccard index of 0.93, a correlation coefficient of 0.99, with a figure-of-merit of 95.99%, while the classifier accuracies for the three DN nets (DN-121, DN-169, and DN-201) were 98%, 98%, and 99% with a loss of ~0.003, ~0.0025, and ~0.002 using 50 epochs, respectively. The mean AUC for all three DN models was 0.99 (p < 0.0001). The COVLIAS 2.0-cXAI showed 80% scans for mean alignment index (MAI) between heatmaps and gold standard, a score of four out of five, establishing the system for clinical settings. Conclusions: The COVLIAS 2.0-cXAI successfully showed a cloud-based explainable AI system for lesion localization in lung CT scans.
ABSTRACT
BACKGROUND: COVLIAS 1.0: an automated lung segmentation was designed for COVID-19 diagnosis. It has issues related to storage space and speed. This study shows that COVLIAS 2.0 uses pruned AI (PAI) networks for improving both storage and speed, wiliest high performance on lung segmentation and lesion localization. METHOD: ology: The proposed study uses multicenter â¼9,000 CT slices from two different nations, namely, CroMed from Croatia (80 patients, experimental data), and NovMed from Italy (72 patients, validation data). We hypothesize that by using pruning and evolutionary optimization algorithms, the size of the AI models can be reduced significantly, ensuring optimal performance. Eight different pruning techniques (i) differential evolution (DE), (ii) genetic algorithm (GA), (iii) particle swarm optimization algorithm (PSO), and (iv) whale optimization algorithm (WO) in two deep learning frameworks (i) Fully connected network (FCN) and (ii) SegNet were designed. COVLIAS 2.0 was validated using "Unseen NovMed" and benchmarked against MedSeg. Statistical tests for stability and reliability were also conducted. RESULTS: Pruning algorithms (i) FCN-DE, (ii) FCN-GA, (iii) FCN-PSO, and (iv) FCN-WO showed improvement in storage by 92.4%, 95.3%, 98.7%, and 99.8% respectively when compared against solo FCN, and (v) SegNet-DE, (vi) SegNet-GA, (vii) SegNet-PSO, and (viii) SegNet-WO showed improvement by 97.1%, 97.9%, 98.8%, and 99.2% respectively when compared against solo SegNet. AUC > 0.94 (p < 0.0001) on CroMed and > 0.86 (p < 0.0001) on NovMed data set for all eight EA model. PAI <0.25 s per image. DenseNet-121-based Grad-CAM heatmaps showed validation on glass ground opacity lesions. CONCLUSIONS: Eight PAI networks that were successfully validated are five times faster, storage efficient, and could be used in clinical settings.
Subject(s)
COVID-19 , Deep Learning , COVID-19/diagnostic imaging , COVID-19 Testing , Humans , Image Processing, Computer-Assisted/methods , Lung/diagnostic imaging , Neural Networks, Computer , Reproducibility of Results , Tomography, X-Ray Computed/methodsABSTRACT
Background: COVID-19 is a disease with multiple variants, and is quickly spreading throughout the world. It is crucial to identify patients who are suspected of having COVID-19 early, because the vaccine is not readily available in certain parts of the world. Methodology: Lung computed tomography (CT) imaging can be used to diagnose COVID-19 as an alternative to the RT-PCR test in some cases. The occurrence of ground-glass opacities in the lung region is a characteristic of COVID-19 in chest CT scans, and these are daunting to locate and segment manually. The proposed study consists of a combination of solo deep learning (DL) and hybrid DL (HDL) models to tackle the lesion location and segmentation more quickly. One DL and four HDL modelsnamely, PSPNet, VGG-SegNet, ResNet-SegNet, VGG-UNet, and ResNet-UNetwere trained by an expert radiologist. The training scheme adopted a fivefold cross-validation strategy on a cohort of 3000 images selected from a set of 40 COVID-19-positive individuals. Results: The proposed variability study uses tracings from two trained radiologists as part of the validation. Five artificial intelligence (AI) models were benchmarked against MedSeg. The best AI model, ResNet-UNet, was superior to MedSeg by 9% and 15% for Dice and Jaccard, respectively, when compared against MD 1, and by 4% and 8%, respectively, when compared against MD 2. Statistical testsnamely, the Mann−Whitney test, paired t-test, and Wilcoxon testdemonstrated its stability and reliability, with p < 0.0001. The online system for each slice was <1 s. Conclusions: The AI models reliably located and segmented COVID-19 lesions in CT scans. The COVLIAS 1.0Lesion lesion locator passed the intervariability test.
ABSTRACT
(1) Background: COVID-19 computed tomography (CT) lung segmentation is critical for COVID lung severity diagnosis. Earlier proposed approaches during 2020-2021 were semiautomated or automated but not accurate, user-friendly, and industry-standard benchmarked. The proposed study compared the COVID Lung Image Analysis System, COVLIAS 1.0 (GBTI, Inc., and AtheroPointTM, Roseville, CA, USA, referred to as COVLIAS), against MedSeg, a web-based Artificial Intelligence (AI) segmentation tool, where COVLIAS uses hybrid deep learning (HDL) models for CT lung segmentation. (2) Materials and Methods: The proposed study used 5000 ITALIAN COVID-19 positive CT lung images collected from 72 patients (experimental data) that confirmed the reverse transcription-polymerase chain reaction (RT-PCR) test. Two hybrid AI models from the COVLIAS system, namely, VGG-SegNet (HDL 1) and ResNet-SegNet (HDL 2), were used to segment the CT lungs. As part of the results, we compared both COVLIAS and MedSeg against two manual delineations (MD 1 and MD 2) using (i) Bland-Altman plots, (ii) Correlation coefficient (CC) plots, (iii) Receiver operating characteristic curve, and (iv) Figure of Merit and (v) visual overlays. A cohort of 500 CROATIA COVID-19 positive CT lung images (validation data) was used. A previously trained COVLIAS model was directly applied to the validation data (as part of Unseen-AI) to segment the CT lungs and compare them against MedSeg. (3) Result: For the experimental data, the four CCs between COVLIAS (HDL 1) vs. MD 1, COVLIAS (HDL 1) vs. MD 2, COVLIAS (HDL 2) vs. MD 1, and COVLIAS (HDL 2) vs. MD 2 were 0.96, 0.96, 0.96, and 0.96, respectively. The mean value of the COVLIAS system for the above four readings was 0.96. CC between MedSeg vs. MD 1 and MedSeg vs. MD 2 was 0.98 and 0.98, respectively. Both had a mean value of 0.98. On the validation data, the CC between COVLIAS (HDL 1) vs. MedSeg and COVLIAS (HDL 2) vs. MedSeg was 0.98 and 0.99, respectively. For the experimental data, the difference between the mean values for COVLIAS and MedSeg showed a difference of <2.5%, meeting the standard of equivalence. The average running times for COVLIAS and MedSeg on a single lung CT slice were ~4 s and ~10 s, respectively. (4) Conclusions: The performances of COVLIAS and MedSeg were similar. However, COVLIAS showed improved computing time over MedSeg.
ABSTRACT
Background: For COVID-19 lung severity, segmentation of lungs on computed tomography (CT) is the first crucial step. Current deep learning (DL)-based Artificial Intelligence (AI) models have a bias in the training stage of segmentation because only one set of ground truth (GT) annotations are evaluated. We propose a robust and stable inter-variability analysis of CT lung segmentation in COVID-19 to avoid the effect of bias. Methodology: The proposed inter-variability study consists of two GT tracers for lung segmentation on chest CT. Three AI models, PSP Net, VGG-SegNet, and ResNet-SegNet, were trained using GT annotations. We hypothesized that if AI models are trained on the GT tracings from multiple experience levels, and if the AI performance on the test data between these AI models is within the 5% range, one can consider such an AI model robust and unbiased. The K5 protocol (training to testing: 80%:20%) was adapted. Ten kinds of metrics were used for performance evaluation. Results: The database consisted of 5000 CT chest images from 72 COVID-19-infected patients. By computing the coefficient of correlations (CC) between the output of the two AI models trained corresponding to the two GT tracers, computing their differences in their CC, and repeating the process for all three AI-models, we show the differences as 0%, 0.51%, and 2.04% (all < 5%), thereby validating the hypothesis. The performance was comparable; however, it had the following order: ResNet-SegNet > PSP Net > VGG-SegNet. Conclusions: The AI models were clinically robust and stable during the inter-variability analysis on the CT lung segmentation on COVID-19 patients.
ABSTRACT
BACKGROUND: COVID-19 lung segmentation using Computed Tomography (CT) scans is important for the diagnosis of lung severity. The process of automated lung segmentation is challenging due to (a) CT radiation dosage and (b) ground-glass opacities caused by COVID-19. The lung segmentation methodologies proposed in 2020 were semi- or automated but not reliable, accurate, and user-friendly. The proposed study presents a COVID Lung Image Analysis System (COVLIAS 1.0, AtheroPoint™, Roseville, CA, USA) consisting of hybrid deep learning (HDL) models for lung segmentation. METHODOLOGY: The COVLIAS 1.0 consists of three methods based on solo deep learning (SDL) or hybrid deep learning (HDL). SegNet is proposed in the SDL category while VGG-SegNet and ResNet-SegNet are designed under the HDL paradigm. The three proposed AI approaches were benchmarked against the National Institute of Health (NIH)-based conventional segmentation model using fuzzy-connectedness. A cross-validation protocol with a 40:60 ratio between training and testing was designed, with 10% validation data. The ground truth (GT) was manually traced by a radiologist trained personnel. For performance evaluation, nine different criteria were selected to perform the evaluation of SDL or HDL lung segmentation regions and lungs long axis against GT. RESULTS: Using the database of 5000 chest CT images (from 72 patients), COVLIAS 1.0 yielded AUC of ~0.96, ~0.97, ~0.98, and ~0.96 (p-value < 0.001), respectively within 5% range of GT area, for SegNet, VGG-SegNet, ResNet-SegNet, and NIH. The mean Figure of Merit using four models (left and right lung) was above 94%. On benchmarking against the National Institute of Health (NIH) segmentation method, the proposed model demonstrated a 58% and 44% improvement in ResNet-SegNet, 52% and 36% improvement in VGG-SegNet for lung area, and lung long axis, respectively. The PE statistics performance was in the following order: ResNet-SegNet > VGG-SegNet > NIH > SegNet. The HDL runs in <1 s on test data per image. CONCLUSIONS: The COVLIAS 1.0 system can be applied in real-time for radiology-based clinical settings.
ABSTRACT
Artificial intelligence (AI) has penetrated the field of medicine, particularly the field of radiology. Since its emergence, the highly virulent coronavirus disease 2019 (COVID-19) has infected over 10 million people, leading to over 500,000 deaths as of July 1st, 2020. Since the outbreak began, almost 28,000 articles about COVID-19 have been published (https://pubmed.ncbi.nlm.nih.gov); however, few have explored the role of imaging and artificial intelligence in COVID-19 patients-specifically, those with comorbidities. This paper begins by presenting the four pathways that can lead to heart and brain injuries following a COVID-19 infection. Our survey also offers insights into the role that imaging can play in the treatment of comorbid patients, based on probabilities derived from COVID-19 symptom statistics. Such symptoms include myocardial injury, hypoxia, plaque rupture, arrhythmias, venous thromboembolism, coronary thrombosis, encephalitis, ischemia, inflammation, and lung injury. At its core, this study considers the role of image-based AI, which can be used to characterize the tissues of a COVID-19 patient and classify the severity of their infection. Image-based AI is more important than ever as the pandemic surges and countries worldwide grapple with limited medical resources for detection and diagnosis.
Subject(s)
Betacoronavirus , Brain Injuries/epidemiology , Coronavirus Infections/epidemiology , Heart Injuries/epidemiology , Pneumonia, Viral/epidemiology , Artificial Intelligence , Betacoronavirus/pathogenicity , Betacoronavirus/physiology , Brain Injuries/classification , Brain Injuries/diagnostic imaging , COVID-19 , COVID-19 Testing , Clinical Laboratory Techniques/methods , Comorbidity , Computational Biology , Coronavirus Infections/classification , Coronavirus Infections/diagnosis , Coronavirus Infections/diagnostic imaging , Deep Learning , Heart Injuries/classification , Heart Injuries/diagnostic imaging , Humans , Machine Learning , Pandemics/classification , Pneumonia, Viral/classification , Pneumonia, Viral/diagnostic imaging , Risk Factors , SARS-CoV-2 , Severity of Illness IndexABSTRACT
BACKGROUND: Status epilepticus is a medical emergency associated with significant mortality and morbidity that requires immediate and effective treatment. OBJECTIVES: (1) To determine whether a particular anticonvulsant is more effective or safer to use in status epilepticus compared to another and compared to placebo.(2) To delineate reasons for disagreement in the literature regarding recommended treatment regimens and to highlight areas for future research. SEARCH METHODS: For the latest update of this review, the following electronic databases were searched on 15/08/2013: the Cochrane Epilepsy Group's Specialized Register, CENTRAL The Cochrane Library July 2013, Issue 7, and MEDLINE (Ovid) 1946 to 15/08/2013. SELECTION CRITERIA: Randomised controlled trials of participants with premonitory, early, established or refractory status epilepticus using a truly random or quasi-random allocation of treatments were included. DATA COLLECTION AND ANALYSIS: Two review authors independently selected trials for inclusion, assessed trial quality and extracted data. MAIN RESULTS: Eighteen studies with 2755 participants were included. Few studies used the same interventions. Intravenous diazepam was better than placebo in reducing the risk of non-cessation of seizures (risk ratio (RR) 0.73, 95% confidence interval (CI) 0.57 to 0.92), requirement for ventilatory support (RR 0.39, 95% CI 0.16 to 0.94), or continuation of status epilepticus requiring use of a different drug or general anaesthesia (RR 0.73, 95% CI 0.57 to 0.92). Intravenous lorazepam was better than placebo for risk of non-cessation of seizures (RR 0.52, 95% CI 0.38 to 0.71) and for risk of continuation of status epilepticus requiring a different drug or general anaesthesia (RR 0.52, 95% CI 0.38 to 0.71). Intravenous lorazepam was better than intravenous diazepam for reducing the risk of non-cessation of seizures (RR 0.64, 95% CI 0.45 to 0.90) and had a lower risk for continuation of status epilepticus requiring a different drug or general anaesthesia (RR 0.63, 95% CI 0.45 to 0.88). Intravenous lorazepam was better than intravenous phenytoin for risk of non-cessation of seizures (RR 0.62, 95% CI 0.45 to 0.86). Diazepam gel was better than placebo gel in reducing the risk of non-cessation of seizures (RR 0.43 95% CI 0.30 to 0.62)For pre-hospital treatment, intramuscular midazolam is at least as effective as (probably more effective than) intravenous lorazepam in control of seizures (RR1.16, 95% CI 1.06 to 1.27) and frequency of hospitalisation (RR 0.88, 95% CI 0.79 to 0.97) or intensive care admissions (RR 0.79, 95% CI 0.65 to 0.96). It was uncertain whether Intravenous valproate was better than intravenous phenytoin in reducing risk of non-cessation of seizures (RR 0.75, 95% CI 0.28 to 2.00). Both levetiracetam and lorazepam were equally effective in aborting seizures (RR 0.97, 95% CI 0.44 to 2.13). Results for other comparisons of anticonvulsant therapies were uncertain due to single studies with few participants.The body of randomised evidence to guide clinical decisions is small. It was uncertain whether any anticonvulsant therapy was better than another in terms of adverse effects, due to few studies and participants identified. The quality of the evidence from the included studies is not strong but appears acceptable. We were unable to make judgements for risk of bias domains incomplete outcome reporting (attrition bias) and selective outcome reporting (selection bias) due to unclear reporting by the study authors. AUTHORS' CONCLUSIONS: Intravenous lorazepam is better than intravenous diazepam or intravenous phenytoin alone for cessation of seizures. Intravenous lorazepam also carries a lower risk of continuation of status epilepticus requiring a different drug or general anaesthesia compared with intravenous diazepam. Both intravenous lorazepam and diazepam are better than placebo for the same outcomes. For pre hospital management, midazolam IM seemed more effective than lorazepam IV for cessation of seizures, frequency of hospitalisation and ICU admissions however,it was unclear whether the risk of recurrence of seizures differed between treatments. The results of other comparisons of anticonvulsant therapies versus each other were also uncertain. Universally accepted definitions of premonitory, early, established and refractory status epilepticus are required. Diazepam gel was better than placebo gel in reducing the risk of non-cessation of seizures. Results for other comparisons of anticonvulsant therapies were uncertain due to single studies with few participants.
Subject(s)
Anticonvulsants/therapeutic use , Status Epilepticus/drug therapy , Diazepam/therapeutic use , Humans , Injections, Intravenous , Lorazepam/therapeutic use , Midazolam/therapeutic use , Phenobarbital/therapeutic use , Phenytoin/therapeutic use , Randomized Controlled Trials as TopicABSTRACT
AIMS: To determine whether a particular anticonvulsant is more effective or safer than another or placebo in patients with status epilepticus, and to summarize the available evidence from randomized controlled trials, and to highlight areas for future research in status epilepticus. METHODS: Randomized controlled trials of participants with premonitory, early, established or refractory status epilepticus using a truly random or quasi-random allocation of treatments were included. RESULTS: Eleven studies with 2017 participants met the inclusion criteria. Lorazepam was better than diazepam for reducing risk of seizure continuation [relative risk (RR) 0.64, 95% confidence interval (CI) 0.45, 0.90] and of requirement of a different drug or general anaesthesia (RR 0.63, 95% CI 0.45, 0.88) with no statistically significant difference in the risk of adverse effects. Lorazepam was better than phenytoin for risk of seizure continuation (RR 0.62, 95% CI 0.45, 0.86). Diazepam 30 mg intrarectal gel was better than 20 mg in premonitory status epilepticus for the risk of seizure continuation (RR 0.39, 95% CI 0.18, 0.86). CONCLUSIONS: Lorazepam is better than diazepam or phenytoin alone for cessation of seizures and carries a lower risk of continuation of status epilepticus requiring a different drug or general anaesthesia. Both lorazepam and diazepam are better than placebo for the same outcomes. In the treatment of premonitory seizures, diazepam 30 mg intrarectal gel is better than 20 mg for cessation of seizures without a statistically significant increase in adverse effects. Universally accepted definitions of premonitory, early, established and refractory status epilepticus are required.
Subject(s)
Anticonvulsants/therapeutic use , Status Epilepticus/drug therapy , Diazepam/therapeutic use , Humans , Lorazepam/therapeutic use , Midazolam/therapeutic use , Phenobarbital/therapeutic use , Phenytoin/therapeutic use , Randomized Controlled Trials as TopicABSTRACT
The association of antiphospholipid antibodies and Takayasu arteritis is very rare and few cases have been reported in the past. Though Takayasu arteritis patients were treated in the past with stenting, there have been no reports of patients with this association being treated with carotid stenting. We present here a young Bahraini female with Takayasu arteritis, primary antiphospholipid antibody syndrome and methylene tetrahydrofolate reductase C 677 T and A 1298 C polymorphism, who was treated with carotid stenting and anticoagulants.
Subject(s)
Antibodies, Antiphospholipid/blood , Methylenetetrahydrofolate Reductase (NADPH2)/genetics , Polymorphism, Single Nucleotide , Takayasu Arteritis/genetics , Adult , Antiphospholipid Syndrome/diagnostic imaging , Antiphospholipid Syndrome/genetics , Antiphospholipid Syndrome/immunology , Aorta, Thoracic/diagnostic imaging , Carotid Artery, Common/diagnostic imaging , Carotid Stenosis/surgery , Coronary Angiography , Cytosine , Female , Humans , Stents , Takayasu Arteritis/diagnostic imaging , Takayasu Arteritis/enzymology , Takayasu Arteritis/immunology , ThymineABSTRACT
Restless legs syndrome (RLS) is a disorder of motor activity with a circadian pattern, occurring frequently in patients with Parkinson's disease (PD). We sought to estimate the prevalence of RLS in Indian PD patients. One hundred twenty-six consecutive PD patients and 128 healthy age- and sex-matched controls were evaluated using a predesigned questionnaire. RLS was present in 10 of 126 cases of PD (7.9%) and 1 of 128 controls (0.8%, P = 0.01). PD patients with RLS were older than those without RLS (63.70 +/- 7.80 years vs. 57.37 +/- 10.04 years; P = 0.05) and had higher prevalence of depression (40% vs. 10.3%; P = 0.023). No demographic factors or factors related to PD correlated with the presence or severity of RLS. RLS is more common among patients with PD than controls. A greater medical recognition of this disorder is needed in view of available effective treatment.