The importance of standardisation - COVID-19 CT & Radiograph Image Data Stock for deep learning purpose.
Ontology highlight
ABSTRACT:
SUBMITTER: Misztal K
PROVIDER: S-EPMC7591316 | biostudies-literature | 2020 Dec
REPOSITORIES: biostudies-literature
Publications
The importance of standardisation - COVID-19 CT & Radiograph Image Data Stock for deep learning purpose.
Misztal Krzysztof K Pocha Agnieszka A Durak-Kozica Martyna M Wątor Michał M Kubica-Misztal Aleksandra A Hartel Marcin M
Computers in biology and medicine 20201028
With the number of affected individuals still growing world-wide, the research on COVID-19 is continuously expanding. The deep learning community concentrates their efforts on exploring if neural networks can potentially support the diagnosis using CT and radiograph images of patients' lungs. The two most popular publicly available datasets for COVID-19 classification are COVID-CT and COVID-19 Image Data Collection. In this work, we propose a new dataset which we call COVID-19 CT & Radiograph Im ...[more]
Similar Datasets
Project description:PurposeTo evaluate the effect of a deep learning-based reconstruction (DLR) method on the conspicuity of hypovascular hepatic metastases on abdominal CT images.Materials and methodsThis retrospective study with institutional review board approval included 58 patients with hypovascular hepatic metastases. A radiologist recorded the standard deviation of attenuation in the paraspinal muscle as the image noise and the contrast-to-noise ratio (CNR). CNR was calculated as region of interest ([ROI]L - ROIT)/N, where ROIL is the mean liver parenchyma attenuation, ROIT, the mean tumor attenuation, and N, the noise. Two other radiologists graded the conspicuity of the liver lesion on a five-point scale where 1 is unidentifiable and 5 is detected without diagnostic compromise. Only the smallest liver lesion in each patient, classified as smaller or larger than 10 mm, was evaluated. The difference between hybrid iterative reconstruction (IR) and DLR images was determined by using a two-sided Wilcoxon signed-rank test.ResultsThe image noise was significantly lower, and the CNR was significantly higher on DLR images than hybrid IR images (median image noise: 19.2 vs 12.8 HU, P < .001; median CNR: tumors < 10 mm: 1.9 vs 2.5; tumors > 10 mm: 1.7 vs 2.2, both P < .001). The scores for liver lesions were significantly higher for DLR images than hybrid IR images (P < .01 for both in tumors smaller or larger than 10 mm).ConclusionDLR improved the quality of abdominal CT images for the evaluation of hypovascular hepatic metastases.© RSNA, 2019Supplemental material is available for this article.
| S-EPMC8017421 | biostudies-literature
Project description:Prognosis of primary pontine hemorrhage (PPH) is important for treatment planning and patient management. However, only few clinical factors were reported to have prognostic value to PPH. Here, we propose a deep learning (DL) model that mines high-dimensional prognostic information from computed tomography (CT) images and combines clinical factors for predicting individualized prognosis of PPH. We proposed a multi-task DL model to learn high-dimensional CT features of hematoma and perihematomal areas for predicting the risk of 30-day mortality, 90-day mortality and 90-day functional outcome of PPH simultaneously. We further explored the combination of the DL model and clinical factors by building a combined model. All the models were trained in a training cohort (n = 219) and tested in an independent testing cohort (n = 35). The DL model achieved area under the curve (AUC) of 0.886, 0.886, and 0.759 in predicting 30-day mortality, 90-day mortality and 90-day functional outcome of PPH in the independent testing cohort, which improved over the previously reported new PPH score and the clinical model. When combining the DL model and clinical factors, the combined model achieved improved performance (AUC = 0.920, 0.941, and 0.894), indicating that DL model mines CT information that complements clinical factors. Through DL visualization technique, we found that the internal structure of hematoma and its expansion to perihematomal regions are important for predicting the prognosis of PPH. This DL model provides an easy-to-use way for predicting individualized prognosis of PPH by mining high-dimensional information from CT images, and showed improvement over clinical factors and present methods.
| S-EPMC9668666 | biostudies-literature
Project description:BackgroundThis study aims to evaluate the impact of a novel deep learning-based image reconstruction (DLIR) algorithm on the image quality in computed tomographic angiography (CTA) for pre-interventional planning of transcatheter aortic valve implantation (TAVI).MethodsWe analyzed 50 consecutive patients (median age 80 years, 25 men) who underwent TAVI planning CT on a 256-dectector-row CT. Images were reconstructed with adaptive statistical iterative reconstruction V (ASIR-V) and DLIR. Intravascular image noise, edge sharpness, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were quantified for ascending aorta, descending aorta, abdominal aorta and iliac arteries. Two readers (one radiologist and one interventional cardiologist) scored task-specific subjective image quality on a five-point scale.ResultsDLIR significantly reduced median image noise by 29-57% at all anatomical locations (all P<0.001). Accordingly, median SNR improved by 44-133% (all P<0.001) and median CNR improved by 44-125% (all P<0.001). DLIR significantly improved subjective image quality for all four pre-specified TAVI-specific tasks (measuring the annulus, assessing valve morphology and calcifications, the coronary ostia, and the suitability of the aorto-iliac access route) for both the radiologist and the interventional cardiologist (P≤0.001). Measurements of the aortic annulus circumference, area and diameter did not differ between ASIR-V and DLIR reconstructions (all P>0.05).ConclusionsDLIR significantly improves objective and subjective image quality in TAVI planning CT compared to a state-of-the-art iterative reconstruction without affecting measurements of the aortic annulus. This may provide an opportunity for further reductions in contrast medium volume in this population.
| S-EPMC9929406 | biostudies-literature
Project description:PurposeHigh-dimensional image features that underlie COVID-19 pneumonia remain opaque. We aim to compare feature engineering and deep learning methods to gain insights into the image features that drive CT-based for COVID-19 pneumonia prediction, and uncover CT image features significant for COVID-19 pneumonia from deep learning and radiomics framework.MethodsA total of 266 patients with COVID-19 and other viral pneumonia with clinical symptoms and CT signs similar to that of COVID-19 during the outbreak were retrospectively collected from three hospitals in China and the USA. All the pneumonia lesions on CT images were manually delineated by four radiologists. One hundred eighty-four patients (n = 93 COVID-19 positive; n = 91 COVID-19 negative; 24,216 pneumonia lesions from 12,001 CT image slices) from two hospitals from China served as discovery cohort for model development. Thirty-two patients (17 COVID-19 positive, 15 COVID-19 negative; 7883 pneumonia lesions from 3799 CT image slices) from a US hospital served as external validation cohort. A bi-directional adversarial network-based framework and PyRadiomics package were used to extract deep learning and radiomics features, respectively. Linear and Lasso classifiers were used to develop models predictive of COVID-19 versus non-COVID-19 viral pneumonia.Results120-dimensional deep learning image features and 120-dimensional radiomics features were extracted. Linear and Lasso classifiers identified 32 high-dimensional deep learning image features and 4 radiomics features associated with COVID-19 pneumonia diagnosis (P < 0.0001). Both models achieved sensitivity > 73% and specificity > 75% on external validation cohort with slight superior performance for radiomics Lasso classifier. Human expert diagnostic performance improved (increase by 16.5% and 11.6% in sensitivity and specificity, respectively) when using a combined deep learning-radiomics model.ConclusionsWe uncover specific deep learning and radiomics features to add insight into interpretability of machine learning algorithms and compare deep learning and radiomics models for COVID-19 pneumonia that might serve to augment human diagnostic performance.
| S-EPMC7581467 | biostudies-literature
Project description:BackgroundAccurate assessment of coronavirus disease 2019 (COVID-19) lung involvement through chest radiograph plays an important role in effective management of the infection. This study aims to develop a two-step feature merging method to integrate image features from deep learning and radiomics to differentiate COVID-19, non-COVID-19 pneumonia and normal chest radiographs (CXR).MethodsIn this study, a deformable convolutional neural network (deformable CNN) was developed and used as a feature extractor to obtain 1,024-dimensional deep learning latent representation (DLR) features. Then 1,069-dimensional radiomics features were extracted from the region of interest (ROI) guided by deformable CNN's attention. The two feature sets were concatenated to generate a merged feature set for classification. For comparative experiments, the same process has been applied to the DLR-only feature set for verifying the effectiveness of feature concatenation.ResultsUsing the merged feature set resulted in an overall average accuracy of 91.0% for three-class classification, representing a statistically significant improvement of 0.6% compared to the DLR-only classification. The recall and precision of classification into the COVID-19 class were 0.926 and 0.976, respectively. The feature merging method was shown to significantly improve the classification performance as compared to using only deep learning features, regardless of choice of classifier (P value <0.0001). Three classes' F1-score were 0.892, 0.890, and 0.950 correspondingly (i.e., normal, non-COVID-19 pneumonia, COVID-19).ConclusionsA two-step COVID-19 classification framework integrating information from both DLR and radiomics features (guided by deep learning attention mechanism) has been developed. The proposed feature merging method has been shown to improve the performance of chest radiograph classification as compared to the case of using only deep learning features.
| S-EPMC9929417 | biostudies-literature
Project description:BackgroundDeep learning using convolutional neural networks (CNN) has achieved significant results in various fields that use images. Deep learning can automatically extract features from data, and CNN extracts image features by convolution processing. We assumed that increasing the image size using interpolation methods would result in an effective feature extraction. To investigate how interpolation methods change as the number of data increases, we examined and compared the effectiveness of data augmentation by inversion or rotation with image augmentation by interpolation when the image data for training were small. Further, we clarified whether image augmentation by interpolation was useful for CNN training. To examine the usefulness of interpolation methods in medical images, we used a Gender01 data set, which is a sex classification data set, on chest radiographs. For comparison of image enlargement using an interpolation method with data augmentation by inversion and rotation, we examined the results of two- and four-fold enlargement using a Bilinear method.ResultsThe average classification accuracy improved by expanding the image size using the interpolation method. The biggest improvement was noted when the number of training data was 100, and the average classification accuracy of the training model with the original data was 0.563. However, upon increasing the image size by four times using the interpolation method, the average classification accuracy significantly improved to 0.715. Compared with the data augmentation by inversion and rotation, the model trained using the Bilinear method showed an improvement in the average classification accuracy by 0.095 with 100 training data and 0.015 with 50,000 training data. Comparisons of the average classification accuracy of the chest X-ray images showed a stable and high-average classification accuracy using the interpolation method.ConclusionTraining the CNN by increasing the image size using the interpolation method is a useful method. In the future, we aim to conduct additional verifications using various medical images to further clarify the reason why image size is important.
| S-EPMC7924688 | biostudies-literature
Project description:Achieving seed germination quality standards poses a real challenge to seed companies as they are compelled to abide by strict certification rules, while having only partial seed separation solutions at their disposal. This discrepancy results with wasteful disqualification of seed lots holding considerable amounts of good seeds and further translates to financial losses and supply chain insecurity. Here, we present the first-ever generic germination prediction technology that is based on deep learning and RGB image data and facilitates seed classification by seed germinability and usability, two facets of germination fate. We show technology competence to render dozens of disqualified seed lots of seven vegetable crops, representing different genetics and production pipelines, industrially appropriate, and to adequately classify lots by utilizing available crop-level image data, instead of lot-specific data. These achievements constitute a major milestone in the deployment of this technology for industrial seed sorting by germination fate for multiple crops.
| S-EPMC8586350 | biostudies-literature
Project description:Urban area mapping is an important application of remote sensing which aims at both estimation and change in land cover under the urban area. A major challenge being faced while analyzing Synthetic Aperture Radar (SAR) based remote sensing data is that there is a lot of similarity between highly vegetated urban areas and oriented urban targets with that of actual vegetation. This similarity between some urban areas and vegetation leads to misclassification of the urban area into forest cover. The present work is a precursor study for the dual-frequency L and S-band NASA-ISRO Synthetic Aperture Radar (NISAR) mission and aims at minimizing the misclassification of such highly vegetated and oriented urban targets into vegetation class with the help of deep learning. In this study, three machine learning algorithms Random Forest (RF), K-Nearest Neighbour (KNN), and Support Vector Machine (SVM) have been implemented along with a deep learning model DeepLabv3+ for semantic segmentation of Polarimetric SAR (PolSAR) data. It is a general perception that a large dataset is required for the successful implementation of any deep learning model but in the field of SAR based remote sensing, a major issue is the unavailability of a large benchmark labeled dataset for the implementation of deep learning algorithms from scratch. In current work, it has been shown that a pre-trained deep learning model DeepLabv3+ outperforms the machine learning algorithms for land use and land cover (LULC) classification task even with a small dataset using transfer learning. The highest pixel accuracy of 87.78% and overall pixel accuracy of 85.65% have been achieved with DeepLabv3+ and Random Forest performs best among the machine learning algorithms with overall pixel accuracy of 77.91% while SVM and KNN trail with an overall accuracy of 77.01% and 76.47% respectively. The highest precision of 0.9228 is recorded for the urban class for semantic segmentation task with DeepLabv3+ while machine learning algorithms SVM and RF gave comparable results with a precision of 0.8977 and 0.8958 respectively.
| S-EPMC8319419 | biostudies-literature
Project description:Background and Objectives: Although reducing the radiation dose level is important during diagnostic computed tomography (CT) applications, effective image quality enhancement strategies are crucial to compensate for the degradation that is caused by a dose reduction. We performed this prospective study to quantify emphysema on ultra-low-dose CT images that were reconstructed using deep learning-based image reconstruction (DLIR) algorithms, and compared and evaluated the accuracies of DLIR algorithms versus standard-dose CT. Materials and Methods: A total of 32 patients were prospectively enrolled, and all underwent standard-dose and ultra-low-dose (120 kVp; CTDIvol < 0.7 mGy) chest CT scans at the same time in a single examination. A total of six image datasets (filtered back projection (FBP) for standard-dose CT, and FBP, adaptive statistical iterative reconstruction (ASIR-V) 50%, DLIR-low, DLIR-medium, DLIR-high for ultra-low-dose CT) were reconstructed for each patient. Image noise values, emphysema indices, total lung volumes, and mean lung attenuations were measured in the six image datasets and compared (one-way repeated measures ANOVA). Results: The mean effective doses for standard-dose and ultra-low-dose CT scans were 3.43 ± 0.57 mSv and 0.39 ± 0.03 mSv, respectively (p < 0.001). The total lung volume and mean lung attenuation of five image datasets of ultra-low-dose CT scans, emphysema indices of ultra-low-dose CT scans reconstructed using ASIR-V 50 or DLIR-low, and the image noise of ultra-low-dose CT scans that were reconstructed using DLIR-low were not different from those of standard-dose CT scans. Conclusions: Ultra-low-dose CT images that were reconstructed using DLIR-low were found to be useful for emphysema quantification at a radiation dose of only 11% of that required for standard-dose CT.
| S-EPMC9317892 | biostudies-literature
Project description:ObjectivesTo compare the dose reduction potential (DRP) of a vendor-agnostic deep learning model (DLM, ClariCT.AI) with that of a vendor-specific deep learning-based image reconstruction algorithm (DLR, TrueFidelity™).MethodsComputed tomography (CT) images of a multi-sized image quality phantom (Mercury v4.0) were acquired under six radiation dose levels (0.48/0.97/1.93/3.87/7.74/15.47 mGy) and were reconstructed using filtered back projection (FBP) and three strength levels of the DLR (low/medium/high). The FBP images were denoised using the DLM. For all DLM and DLR images, the detectability index (d') (a task-based detection performance metric) was obtained, under various combinations of three target sizes (10/5/1 mm), five inlets (CT value difference with the background; -895/50/90/335/1000 HU), five phantom diameters (36/31/26/21/16 cm), and six radiation dose levels. Dose reduction potential (DRP) measures the dose reduction made by using DLM or DLR, while yielding d' equivalent to that of FBP at full dose.ResultsThe DRPs of the DLM, DLR-low, DLR-medium, and DLR-high were 86% (81-88%), 60% (46-67%), 76% (60-81%), and 87% (78-92%), respectively. For 10-mm targets, the DRP of the DLM (87%) was higher than that of all DLR algorithms (58-86%). However, for smaller targets (5 mm/1 mm), the DRPs of the DLR-high (89/88%) were greater than those of the DLM (87/84%).ConclusionThe dose reduction potential of the vendor-agnostic DLM was shown to be comparable to that of the vendor-specific DLR at high strength and superior to those of the DLRs at medium and low strengths.Key points• DRP of the vendor-agnostic model was comparable to that of high-strength vendor-specific model and superior to those of medium- and low-strength models. • Under various radiation dose levels, the deep learning model shows higher detectability indexes compared to FBP.
| S-EPMC8364308 | biostudies-literature