A deep learning algorithm for automatic detection and classification of acute intracranial hemorrhages in head CT scans.
Ontology highlight
ABSTRACT:
SUBMITTER: Wang X
PROVIDER: S-EPMC8377493 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundHighly accurate detection of intracranial hemorrhages (ICH) on head computed tomography (HCT) scans can prove challenging at high-volume centers. This study aimed to determine the number of additional ICHs detected by an artificial intelligence (AI) algorithm and to evaluate reasons for erroneous results at a level I trauma center with teleradiology services.MethodsIn a retrospective multi-center cohort study, consecutive emergency non-contrast HCT scans were analyzed by a commercially available ICH detection software (AIDOC, Tel Aviv, Israel). Discrepancies between AI analysis and initial radiology report (RR) were reviewed by a blinded neuroradiologist to determine the number of additional ICHs detected and evaluate reasons leading to errors.Results4946 HCT (05/2020-09/2020) from 18 hospitals were included in the analysis. 205 reports (4.1%) were classified as hemorrhages by both radiology report and AI. Out of a total of 162 (3.3%) discrepant reports, 62 were confirmed as hemorrhages by the reference neuroradiologist. 33 ICHs were identified exclusively via RRs. The AI algorithm detected an additional 29 instances of ICH, missed 12.4% of ICH and overcalled 1.9%; RRs missed 10.9% of ICHs and overcalled 0.2%. Many of the ICHs missed by the AI algorithm were located in the subarachnoid space (42.4%) and under the calvaria (48.5%). 85% of ICHs missed by RRs occurred outside of regular working-hours. Calcifications (39.3%), beam-hardening artifacts (18%), tumors (15.7%), and blood vessels (7.9%) were the most common reasons for AI overcalls. ICH size, image quality, and primary examiner experience were not found to be significantly associated with likelihood of incorrect AI results.ConclusionComplementing human expertise with AI resulted in a 12.2% increase in ICH detection. The AI algorithm overcalled 1.9% HCT.Trial registrationGerman Clinical Trials Register (DRKS-ID: DRKS00023593).
| S-EPMC8629230 | biostudies-literature
Project description:BackgroundPattern of emphysema at chest CT, scored visually by using the Fleischner Society system, is associated with physiologic impairment and mortality risk.PurposeTo determine whether participant-level emphysema pattern could predict impairment and mortality when classified by using a deep learning method.Materials and MethodsThis retrospective analysis of Genetic Epidemiology of COPD (COPDGene) study participants enrolled between 2007 and 2011 included those with baseline CT, visual emphysema scores, and survival data through 2018. Participants were partitioned into nonoverlapping sets of 2407 for algorithm training, 100 for validation and parameter tuning, and 7143 for testing. A deep learning algorithm using convolutional neural network and long short-term memory architectures was trained to classify pattern of emphysema according to Fleischner criteria. Deep learning scores were compared with visual scores and clinical parameters including pulmonary function tests. Cox proportional hazard models were used to evaluate relationships between emphysema scores and survival. The algorithm was also tested by using CT and clinical data in 1962 participants enrolled in the Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-points (ECLIPSE) study.ResultsA total of 7143 COPDGene participants (mean age ± standard deviation, 59.8 years ± 8.9; 3734 men and 3409 women) were evaluated. Deep learning emphysema classifications were associated with impaired pulmonary function tests, 6-minute walk distance, and St George's Respiratory Questionnaire at univariate analysis (P < .001 for each). Testing in the ECLIPSE cohort showed similar associations (P < .001). In the COPDGene test cohort, deep learning emphysema classification improved the fit of linear mixed models in the prediction of these clinical parameters compared with visual scoring (P < .001). Compared with participants without emphysema, mortality was greater in participants classified by the deep learning algorithm as having any grade of emphysema (adjusted hazard ratios were 1.5, 1.7, 2.9, 5.3, and 9.7, respectively, for trace, mild, moderate, confluent, and advanced destructive emphysema; P < .05).ConclusionDeep learning automation of the Fleischner grade of emphysema at chest CT is associated with clinical measures of pulmonary insufficiency and the risk of mortality.© RSNA, 2019Online supplemental material is available for this article.
| S-EPMC6996603 | biostudies-literature
Project description:COVID-19 is an infectious and contagious virus. As of this writing, more than 160 million people have been infected since its emergence, including more than 125,000 in Algeria. In this work, We first collected a dataset of 4986 COVID and non-COVID images confirmed by RT-PCR tests at Tlemcen hospital in Algeria. Then we performed a transfer learning on deep learning models that got the best results on the ImageNet dataset, such as DenseNet121, DenseNet201, VGG16, VGG19, Inception Resnet-V2, and Xception, in order to conduct a comparative study. Therefore, We have proposed an explainable model based on the DenseNet201 architecture and the GradCam explanation algorithm to detect COVID-19 in chest CT images and explain the output decision. Experiments have shown promising results and proven that the introduced model can be beneficial for diagnosing and following up patients with COVID-19.
| S-EPMC8455169 | biostudies-literature
Project description:Objective:To demonstrate and test the validity of a novel deep-learning-based system for the automated detection of pulmonary nodules. Materials and Methods:The proposed system uses 2 3D deep learning models, 1 for each of the essential tasks of computer-aided nodule detection: candidate generation and false positive reduction. A total of 888 scans from the LIDC-IDRI dataset were used for training and evaluation. Results:Results for candidate generation on the test data indicated a detection rate of 94.77% with 30.39 false positives per scan, while the test results for false positive reduction exhibited a sensitivity of 94.21% with 1.789 false positives per scan. The overall system detection rate on the test data was 89.29% with 1.789 false positives per scan. Discussion:An extensive and rigorous validation was conducted to assess the performance of the proposed system. The system demonstrated a novel combination of 3D deep neural network architectures and demonstrates the use of deep learning for both candidate generation and false positive reduction to be evaluated with a substantial test dataset. The results strongly support the ability of deep learning pulmonary nodule detection systems to generalize to unseen data. The source code and trained model weights have been made available. Conclusion:A novel deep-neural-network-based pulmonary nodule detection system is demonstrated and validated. The results provide comparison of the proposed deep-learning-based system over other similar systems based on performance.
| S-EPMC6188511 | biostudies-literature
Project description:PurposeThe global noise (GN) algorithm has been previously introduced as a method for automatic noise measurement in clinical CT images. The accuracy of the GN algorithm has been assessed in abdomen CT examinations, but not in any other body part until now. This work assesses the GN algorithm accuracy in automatic noise measurement in head CT examinations.MethodsA publicly available image dataset of 99 head CT examinations was used to evaluate the accuracy of the GN algorithm in comparison to reference noise values. Reference noise values were acquired using a manual noise measurement procedure. The procedure used a consistent instruction protocol and multiple observers to mitigate the influence of intra- and interobserver variation, resulting in precise reference values. Optimal GN algorithm parameter values were determined. The GN algorithm accuracy and the corresponding statistical confidence interval were determined. The GN measurements were compared across the six different scan protocols used in this dataset. The correlation of GN to patient head size was also assessed using a linear regression model, and the CT scanner's X-ray beam quality was inferred from the model fit parameters.ResultsAcross all head CT examinations in the dataset, the range of reference noise was 2.9-10.2 HU. A precision of ±0.33 HU was achieved in the reference noise measurements. After optimization, the GN algorithm had a RMS error 0.34 HU corresponding to a percent RMS error of 6.6%. The GN algorithm had a bias of +3.9%. Statistically significant differences in GN were detected in 11 out of the 15 different pairs of scan protocols. The GN measurements were correlated with head size with a statistically significant regression slope parameter (p < 10-7 ). The CT scanner X-ray beam quality estimated from the slope parameter was 3.5 cm water HVL (2.8-4.8 cm 95% CI).ConclusionThe GN algorithm was validated for application in head CT examinations. The GN algorithm was accurate in comparison to reference manual measurement, with errors comparable to interobserver variation in manual measurement. The GN algorithm can detect noise differences in examinations performed on different scanner models or using different scan protocols. The trend in GN across patients of different head sizes closely follows that predicted by a physical model of X-ray attenuation.
| S-EPMC9291315 | biostudies-literature
Project description:Rationale and objectivesTo develop and evaluate an automated machine learning (ML) algorithm for segmenting the paraspinous muscles on chest computed tomography (CT) scans to evaluate for presence of sarcopenia.Materials and methodsA convolutional neural network based on the U-Net architecture was trained to perform muscle segmentation on a dataset of 1875 single slice CT images and was tested on 209 CT images of participants in the National Lung Screening Trial. Low-dose, noncontrast CT examinations were obtained at 33 clinical sites, using scanners from four manufacturers. The study participants had a mean age of 71.6 years (range, 70-74 years). Ground truth was obtained by manually segmenting the left paraspinous muscle at the level of the T12 vertebra. Muscle cross-sectional area (CSA) and muscle attenuation (MA) were recorded. Comparison between the ML algorithm and ground truth measures of muscle CSA and MA were obtained using Dice similarity coefficients and Pearson correlations.ResultsCompared to ground truth segmentation, the ML algorithm achieved median (standard deviation) Dice scores of 0.94 (0.04) in the test set. Mean (SD) muscle CSA was 14.3 (3.6) cm2 for ground truth and 13.7 (3.5) cm2 for ML segmentation. Mean (SD) MA was 41.6 (7.6) Hounsfield units (HU) for ground truth and 43.5 (7.9) HU for ML segmentation. There was high correlation between ML algorithm and ground truth for muscle CSA (r2 = 0.86; p < 0.0001) and MA (r2 = 0.95; p < 0.0001).ConclusionThe ML algorithm for measurement of paraspinous muscles compared favorably to manual ground truth measurements in the NLST. The algorithm generalized well to a heterogeneous set of low-dose CT images and may be capable of automated quantification of muscle metrics to screen for sarcopenia on routine chest CT examinations.
| S-EPMC6878160 | biostudies-literature
Project description:Tumor histology is an important predictor of therapeutic response and outcomes in lung cancer. Tissue sampling for pathologist review is the most reliable method for histology classification, however, recent advances in deep learning for medical image analysis allude to the utility of radiologic data in further describing disease characteristics and for risk stratification. In this study, we propose a radiomics approach to predicting non-small cell lung cancer (NSCLC) tumor histology from non-invasive standard-of-care computed tomography (CT) data. We trained and validated convolutional neural networks (CNNs) on a dataset comprising 311 early-stage NSCLC patients receiving surgical treatment at Massachusetts General Hospital (MGH), with a focus on the two most common histological types: adenocarcinoma (ADC) and Squamous Cell Carcinoma (SCC). The CNNs were able to predict tumor histology with an AUC of 0.71(p = 0.018). We also found that using machine learning classifiers such as k-nearest neighbors (kNN) and support vector machine (SVM) on CNN-derived quantitative radiomics features yielded comparable discriminative performance, with AUC of up to 0.71 (p = 0.017). Our best performing CNN functioned as a robust probabilistic classifier in heterogeneous test sets, with qualitatively interpretable visual explanations to its predictions. Deep learning based radiomics can identify histological phenotypes in lung cancer. It has the potential to augment existing approaches and serve as a corrective aid for diagnosticians.
| S-EPMC7943565 | biostudies-literature
Project description:PurposeTo develop a fully automated algorithm for spleen segmentation and to assess the performance of this algorithm in a large dataset.Materials and methodsIn this retrospective study, a three-dimensional deep learning network was developed to segment the spleen on thorax-abdomen CT scans. Scans were extracted from patients undergoing oncologic treatment from 2014 to 2017. A total of 1100 scans from 1100 patients were used in this study, and 400 were selected for development of the algorithm. For testing, a dataset of 50 scans was annotated to assess the segmentation accuracy and was compared against the splenic index equation. In a qualitative observer experiment, an enriched set of 100 scan-pairs was used to evaluate whether the algorithm could aid a radiologist in assessing splenic volume change. The reference standard was set by the consensus of two other independent radiologists. A Mann-Whitney U test was conducted to test whether there was a performance difference between the algorithm and the independent observer.ResultsThe algorithm and the independent observer obtained comparable Dice scores (P = .834) on the test set of 50 scans of 0.962 and 0.964, respectively. The radiologist had an agreement with the reference standard in 81% (81 of 100) of the cases after a visual classification of volume change, which increased to 92% (92 of 100) when aided by the algorithm.ConclusionA segmentation method based on deep learning can accurately segment the spleen on CT scans and may help radiologists to detect abnormal splenic volumes and splenic volume changes.Supplemental material is available for this article.© RSNA, 2020.
| S-EPMC8082410 | biostudies-literature
Project description:PurposeTo develop a deep learning model that segments intracranial structures on head CT scans.Materials and methodsIn this retrospective study, a primary dataset containing 62 normal noncontrast head CT scans from 62 patients (mean age, 73 years; age range, 27-95 years) acquired between August and December 2018 was used for model development. Eleven intracranial structures were manually annotated on the axial oblique series. The dataset was split into 40 scans for training, 10 for validation, and 12 for testing. After initial training, eight model configurations were evaluated on the validation dataset and the highest performing model was evaluated on the test dataset. Interobserver variability was reported using multirater consensus labels obtained from the test dataset. To ensure that the model learned generalizable features, it was further evaluated on two secondary datasets containing 12 volumes with idiopathic normal pressure hydrocephalus (iNPH) and 30 normal volumes from a publicly available source. Statistical significance was determined using categorical linear regression with P < .05.ResultsOverall Dice coefficient on the primary test dataset was 0.84 ± 0.05 (standard deviation). Performance ranged from 0.96 ± 0.01 (brainstem and cerebrum) to 0.74 ± 0.06 (internal capsule). Dice coefficients were comparable to expert annotations and exceeded those of existing segmentation methods. The model remained robust on external CT scans and scans demonstrating ventricular enlargement. The use of within-network normalization and class weighting facilitated learning of underrepresented classes.ConclusionAutomated segmentation of CT neuroanatomy is feasible with a high degree of accuracy. The model generalized to external CT scans as well as scans demonstrating iNPH.Supplemental material is available for this article.© RSNA, 2020.
| S-EPMC8082409 | biostudies-literature
Project description:Human machine interfaces that can track head motion will result in advances in physical rehabilitation, improved augmented reality/virtual reality systems, and aid in the study of human behavior. This paper presents a head position monitoring and classification system using thin flexible strain sensing threads placed on the neck of an individual. A wireless circuit module consisting of impedance readout circuitry and a Bluetooth module records and transmits strain information to a computer. A data processing algorithm for motion recognition provides near real-time quantification of head position. Incoming data is filtered, normalized and divided into data segments. A set of features is extracted from each data segment and employed as input to nine classifiers including Support Vector Machine, Naive Bayes and KNN for position prediction. A testing accuracy of around 92% was achieved for a set of nine head orientations. Results indicate that this human machine interface platform is accurate, flexible, easy to use, and cost effective.
| S-EPMC7846730 | biostudies-literature