ABSTRACT
Architectural distortion (AD) is one of the most common findings on mammograms, and it may represent not only cancer but also a lesion such as a radial scar that may have an associated cancer. AD accounts for 18-45% missed cancer, and the positive predictive value of AD is approximately 74.5%. Early detection of AD leads to early diagnosis and treatment of the cancer and improves the overall prognosis. However, detection of AD is a challenging task. In this work, we propose a new approach for detecting architectural distortion in mammography images by combining preprocessing methods and a novel structure fusion attention model. The proposed structure-focused weighted orientation preprocessing method is composed of the original image, the architecture enhancement map, and the weighted orientation map, highlighting suspicious AD locations. The proposed structure fusion attention model captures the information from different channels and outperforms other models in terms of false positives and top sensitivity, which refers to the maximum sensitivity that a model can achieve under the acceptance of the highest number of false positives, reaching 0.92 top sensitivity with only 0.6590 false positive per image. The findings suggest that the combination of preprocessing methods and a novel network architecture can lead to more accurate and reliable AD detection. Overall, the proposed approach offers a novel perspective on detecting ADs, and we believe that our method can be applied to clinical settings in the future, assisting radiologists in the early detection of ADs from mammography, ultimately leading to early treatment of breast cancer patients.
ABSTRACT
BACKGROUND: Predictive analytics is gaining popularity as an aid to treatment planning for patients with bone metastases, whose expected survival should be considered. Decreased psoas muscle area (PMA), a morphometric indicator of suboptimal nutritional status, has been associated with mortality in various cancers, but never been integrated into current survival prediction algorithms (SPA) for patients with skeletal metastases. This study investigates whether decreased PMA predicts worse survival in patients with extremity metastases and whether incorporating PMA into three modern SPAs (PATHFx, SORG-NG, and SORG-MLA) improves their performance. METHODS: One hundred eighty-five patients surgically treated for long-bone metastases between 2014 and 2019 were divided into three PMA tertiles (small, medium, and large) based on their psoas size on CT. Kaplan-Meier, multivariable regression, and Cox proportional hazards analyses were employed to compare survival between tertiles and examine factors associated with mortality. Logistic regression analysis was used to assess whether incorporating adjusted PMA values enhanced the three SPAs' discriminatory abilities. The clinical utility of incorporating PMA into these SPAs was evaluated by decision curve analysis (DCA). RESULTS: Patients with small PMA had worse 90-day and 1-year survival after surgery (log-rank test p < 0.001). Patients in the large PMA group had a higher chance of surviving 90 days (odds ratio, OR, 3.72, p = 0.02) and 1 year than those in the small PMA group (OR 3.28, p = 0.004). All three SPAs had increased AUC after incorporation of adjusted PMA. DCA indicated increased net benefits at threshold probabilities >0.5 after the addition of adjusted PMA to these SPAs. CONCLUSIONS: Decreased PMA on CT is associated with worse survival in surgically treated patients with extremity metastases, even after controlling for three contemporary SPAs. Physicians should consider the additional prognostic value of PMA on survival in patients undergoing consideration for operative management due to extremity metastases.
Subject(s)
Bone Neoplasms , Psoas Muscles , Humans , Psoas Muscles/diagnostic imaging , Retrospective Studies , PrognosisABSTRACT
Glyphosate (GLY) is a widely used herbicide worldwide, particularly in cultivating genetically modified soybeans resistant to GLY. However, routine multi-residue analysis does not include GLY due to the complexity of soybean matrix components that can interfere with the analysis. This study presented the development of an aptamer-based chemiluminescence (Apt-CL) sensor for rapidly screening GLY pesticide residue in soybeans. The GLY-binding aptamer (GBA) was developed to bind to GLY specifically, and the remaining unbound aptamers were adsorbed onto gold nanoparticles (AuNPs). The signal was in the form of luminol-H2O2 emission, catalyzed by the aggregation of AuNPs in a chemiluminescent reaction arising from the GLY-GBA complex. The outcomes demonstrated a robust linear relationship between the CL intensity of GLY-GBA and the GLY concentration. In the specificity test of the GBA, only GLY and Profenofos were distinguished among the fifteen tested pesticides. Furthermore, the Apt-CL sensor was conducted to determine GLY residue in organic soybeans immersed in GLY as a real sample, and an optimal linear concentration range for detection after extraction was found to be between 0.001 and 10 mg/L. The Apt-CL sensor exploits the feasibility of real-time pesticide screening in food safety.
Subject(s)
Aptamers, Nucleotide , Biosensing Techniques , Metal Nanoparticles , Pesticide Residues , Glycine max , Metal Nanoparticles/chemistry , Aptamers, Nucleotide/chemistry , Gold/chemistry , Glyphosate , Luminescence , Hydrogen Peroxide/chemistry , Luminescent MeasurementsABSTRACT
OBJECTIVES: A deep learning (DL) model using image data from pretreatment [ 18 F]fluorodeoxyglucose ([ 18 F] FDG)-PET or computed tomography (CT) augmented with a novel imaging augmentation approach was developed for the early prediction of distant metastases in patients with locally advanced uterine cervical cancer. METHODS: This study used baseline [18F]FDG-PET/CT images of newly diagnosed uterine cervical cancer patients. Data from 186 to 25 patients were analyzed for training and validation cohort, respectively. All patients received chemoradiotherapy (CRT) and follow-up. PET and CT images were augmented by using three-dimensional techniques. The proposed model employed DL to predict distant metastases. Receiver operating characteristic (ROC) curve analysis was performed to measure the model's predictive performance. RESULTS: The area under the ROC curves of the training and validation cohorts were 0.818 and 0.830 for predicting distant metastasis, respectively. In the training cohort, the sensitivity, specificity, and accuracy were 80.0%, 78.0%, and 78.5%, whereas, the sensitivity, specificity, and accuracy for distant failure were 73.3%, 75.5%, and 75.2% in the validation cohort, respectively. CONCLUSION: Through the use of baseline [ 18 F]FDG-PET/CT images, the proposed DL model can predict the development of distant metastases for patients with locally advanced uterine cervical cancer treatment by CRT. External validation must be conducted to determine the model's predictive performance.
Subject(s)
Deep Learning , Uterine Cervical Neoplasms , Female , Humans , Positron Emission Tomography Computed Tomography/methods , Fluorodeoxyglucose F18 , Uterine Cervical Neoplasms/pathology , Radiopharmaceuticals , Chemoradiotherapy , Positron-Emission TomographyABSTRACT
AIMS: Risk stratification of atypical ductal hyperplasia (ADH) and ductal carcinoma in situ (DCIS), diagnosed using breast biopsy, has great clinical significance. Clinical trials are currently exploring the possibility of active surveillance for low-risk lesions, whereas axillary lymph node staging may be considered during surgical planning for high-risk lesions. We aimed to develop a machine-learning algorithm based on whole-slide images of breast biopsy specimens and clinical information to predict the risk of upstaging to invasive breast cancer after wide excision. METHODS AND RESULTS: Patients diagnosed with ADH/DCIS on breast biopsy were included in this study, comprising 592 (740 slides) and 141 (198 slides) patients in the development and independent testing cohorts, respectively. Histological grading of the lesions was independently evaluated by two pathologists. Clinical information, including biopsy method, lesion size, and Breast Imaging Reporting and Data System (BI-RADS) classification of ultrasound and mammograms, were collected. Deep DCIS consisted of three deep neural networks to evaluate nuclear grade, necrosis, and stromal reactivity. Deep DCIS output comprised five parameters: total patches, lesion extent, Deep Grade, Deep Necrosis, and Deep Stroma. Deep DCIS highly correlated with the pathologists' evaluations of both slide- and patient-level labels. All five parameters of Deep DCIS were significantly associated with upstaging to invasive carcinoma in subsequent wide excisional specimens. Using multivariate logistic regression, Deep DCIS predicted upstaging to invasive carcinoma with an area under the curve (AUC) of 0.81, outperforming pathologists' evaluation (AUC, 0.71 and 0.69). After including clinical and hormone receptor status information, performance further improved (AUC, 0.87). This combined model retained its predictive power in two subgroup analyses: the first subgroup included unequivocal DCIS (excluding cases of ADH and DCIS suspicious for microinvasion) (AUC, 0.83), while the second excluded cases of high-grade DCIS (AUC, 0.81). The model was validated in an independent testing cohort (AUC, 0.81). CONCLUSION: This study demonstrated that deep-learning models can refine histological evaluation of ADH and DCIS on breast biopsies, which may help guide future treatment planning.
Subject(s)
Breast Neoplasms , Carcinoma, Ductal, Breast , Carcinoma, Intraductal, Noninfiltrating , Deep Learning , Humans , Female , Carcinoma, Intraductal, Noninfiltrating/pathology , Breast/pathology , Breast Neoplasms/pathology , Biopsy , Necrosis/pathology , Carcinoma, Ductal, Breast/pathology , Retrospective Studies , Hyperplasia/pathologyABSTRACT
OBJECTIVES: To predict KRAS mutation in rectal cancer (RC) through computer vision of [18F]fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/computed tomography (CT) by using metric learning (ML). METHODS: This study included 160 patients with RC who had undergone preoperative PET/CT. KRAS mutation was identified through polymerase chain reaction analysis. This model combined ML with the deep-learning framework to analyze PET data with or without CT images. The Batch Balance Wrapper framework and K-fold cross-validation were employed during the learning process. A receiver operating characteristic (ROC) curve analysis was performed to assess the model's predictive performance. RESULTS: Genetic alterations in KRAS were identified in 82 (51%) tumors. Both PET and CT images were used, and the proposed model had an area under the ROC curve of 0.836 for its ability to predict a mutation status. The sensitivity, specificity, and accuracy were 75.3%, 79.3%, and 77.5%, respectively. When PET images alone were used, the area under the curve was 0.817, whereas the sensitivity, specificity, and accuracy were 73.2%, 79.6%, and 76.2%, respectively. CONCLUSIONS: The ML model presented herein revealed that baseline 18F-FDG PET/CT images could provide supplemental information to determine KRAS mutation in RC. Additional studies are required to maximize the predictive accuracy. ADVANCES IN KNOWLEDGE: The results of the ML model presented herein indicate that baseline 18F-FDG PET/CT images could provide supplemental information for determining KRAS mutation in RC.The predictive accuracy of the model was 77.5% when both image types were used and 76.2% when PET images alone were used. Additional studies are required to maximize the predictive accuracy.
Subject(s)
Positron Emission Tomography Computed Tomography , Rectal Neoplasms , Humans , Fluorodeoxyglucose F18 , Proto-Oncogene Proteins p21(ras)/genetics , Rectal Neoplasms/diagnostic imaging , Rectal Neoplasms/genetics , Mutation , Positron-Emission Tomography/methods , RadiopharmaceuticalsABSTRACT
INTRODUCTION: The current ultrasound scan classification system for thyroid nodules is time-consuming, labor-intensive, and subjective. Artificial intelligence (AI) has been shown to increase the accuracy of predicting the malignancy rate of thyroid nodules. This study aims to demonstrate the state-of-the-art Swin Transformer to classify thyroid nodules. MATERIALS AND METHODS: Ultrasound images were collected prospectively from patients who received fine needle aspiration biopsy for thyroid nodules from January 2016 to June 2021. One hundred thirty-nine patients with malignant thyroid nodules were enrolled, while 235 patients with benign nodules served as controls. Images were fed to Swin-T and ResNeSt50 models to classify the thyroid nodules. RESULTS: Patients with malignant nodules were younger and more likely male compared to those with benign nodules. The average sensitivity and specificity of Swin-T were 82.46% and 84.29%, respectively. The average sensitivity and specificity of ResNeSt50 were 72.51% and 77.14%, respectively. Receiver operating characteristics analysis revealed that the area under the curve of Swin-T was higher (AUC=0.91) than that of ResNeSt50 (AUC=0.82). The McNemar test evaluating the performance of these models showed that Swin-T had significantly better performance than ResNeSt50.Swin-T classifier can be a useful tool in helping shared decision-making between physicians and patients with thyroid nodules, particularly in those with high-risk characteristics of sonographic patterns.
Subject(s)
Deep Learning , Thyroid Neoplasms , Thyroid Nodule , Humans , Male , Thyroid Nodule/diagnostic imaging , Thyroid Nodule/pathology , Artificial Intelligence , Sensitivity and Specificity , Ultrasonography/methods , Thyroid Neoplasms/pathologyABSTRACT
PURPOSE: This study aimed to evaluate the clinical usefulness of a deep learning-based computer-aided detection (CADe) system for breast ultrasound. METHODS: The set of 88 training images was expanded to 14,000 positive images and 50,000 negative images. The CADe system was trained to detect lesions in real- time using deep learning with an improved model of YOLOv3-tiny. Eighteen readers evaluated 52 test image sets with and without CADe. Jackknife alternative free-response receiver operating characteristic analysis was used to estimate the effectiveness of this system in improving lesion detection. RESULT: The area under the curve (AUC) for image sets was 0.7726 with CADe and 0.6304 without CADe, with a 0.1422 difference, indicating that with CADe was significantly higher than that without CADe (p < 0.0001). The sensitivity per case was higher with CADe (95.4%) than without CADe (83.7%). The specificity of suspected breast cancer cases with CADe (86.6%) was higher than that without CADe (65.7%). The number of false positives per case (FPC) was lower with CADe (0.22) than without CADe (0.43). CONCLUSION: The use of a deep learning-based CADe system for breast ultrasound by readers significantly improved their reading ability. This system is expected to contribute to highly accurate breast cancer screening and diagnosis.
Subject(s)
Breast Neoplasms , Deep Learning , Humans , Female , Breast Neoplasms/diagnostic imaging , ROC Curve , ComputersABSTRACT
OBJECTIVES: To mitigate the shortage of homograft sources, the use of handmade trileaflet expanded polytetrafluoroethylene valves in pulmonary valve replacement has shown excellent results from multicentre studies conducted in Japan. However, world-wide data outside Japan are relatively insufficient. This study presents the long-term results of a single surgeon's use of flipped-back trileaflet method in a 10-year case series. METHODS: We have developed an efficient way to make a trileaflet-valved conduit utilizing flipped-back method for pulmonary valve replacement and have employed the technique since 2011. Retrospective data were studied between October 2010 and January 2020. Echocardiography, electrocardiogram, Pro-Brain Natriuretic Peptide and Magnetic Resonance Imaging data were analysed. RESULTS: Fifty-five patients were reviewed and median follow-up duration was 2.9 years. The majority of diagnoses was Tetralogy of Fallot (n = 41), and these patients subsequently underwent secondary pulmonary valve replacement at a median age of 15.6 years. Survival was 92.7% with the longest follow-up period being 10 years. There was no need for reoperation, and freedom from reintervention was 98.0% at 10 years. There were 4 deaths (3 in-hospital and 1 outpatient). One patient eventually received transcatheter pulmonary valve implantation. Postoperative echocardiography showed mild or less pulmonary stenosis and pulmonary regurgitation degree in 92.2% and 92.0% of patients, respectively. Comparable magnetic resonance imaging data (n = 25) showed significant reduction in right ventricular volumes but not in ejection fractions. CONCLUSIONS: Our series showed satisfactory long-term function of handmade flipped-back trileaflet-valved conduit used in our patients. The simple design is efficiently reproducible without complex fabrication process.
Subject(s)
Heart Valve Prosthesis Implantation , Heart Valve Prosthesis , Pulmonary Valve , Ventricular Outflow Obstruction , Humans , Adolescent , Pulmonary Valve/surgery , Polytetrafluoroethylene , Retrospective Studies , Prosthesis Design , Treatment Outcome , Ventricular Outflow Obstruction/surgeryABSTRACT
BACKGROUND AND OBJECTIVE: Lung cancer has the highest cancer-related mortality worldwide, and lung nodule usually presents with no symptom. Low-dose computed tomography (LDCT) was an important tool for lung cancer detection and diagnosis. It provided a complete three-dimensional (3-D) chest image with a high resolution.Recently, convolutional neural network (CNN) had flourished and been proven the CNN-based computer-aided diagnosis (CADx) system could extract the features and help radiologists to make a preliminary diagnosis. Therefore, a 3-D ResNeXt-based CADx system was proposed to assist radiologists for diagnosis in this study. METHODS: The proposed CADx system consists of image preprocessing and a 3-D CNN-based classification model for pulmonary nodule classification. First, the image preprocessing was executed to generate the normalized volumn of interest (VOI) only including nodule information and a few surrounding tissues. Then, the extracted VOI was forwarded to the 3-D nodule classification model. In the classification model, the RestNext was employed as the backbone and the attention scheme was embedded to focus on the important features. Moreover, a multi-level feature fusion network incorporating feature information of different scales was used to enhance the prediction accuracy of small malignant nodules. Finally, a hybrid loss based on channel optimization which make the network learn more detailed information was empolyed to replace a binary cross-entropy (BCE) loss. RESULTS: In this research, there were a total of 880 low-dose CT images including 440 benign and 440 malignant nodules from the American National Lung Screening Trial (NLST) for system evaluation. The results showed that our system could achieve the accuracy of 85.3%, the sensitivity of 86.8%, the specificity of 83.9%, and the area-under-curve (AUC) value was 0.9042. It was confirmed that the designed system had a good diagnostic ability. CONCLUSION: In this study, a CADx composed of the image preprocessing and a 3-D nodule classification model with attention scheme, feature fusion, and hybrid loss was proposed for pulmonary nodule classification in LDCT. The results indicated that the proposed CADx system had potential for achieving high performance in classifying lung nodules as benign and malignant.
Subject(s)
Lung Neoplasms , Solitary Pulmonary Nodule , Humans , Neural Networks, Computer , Lung/pathology , Lung Neoplasms/diagnostic imaging , Lung Neoplasms/pathology , Tomography, X-Ray Computed/methods , Diagnosis, Computer-Assisted/methods , Solitary Pulmonary Nodule/diagnostic imaging , Radiographic Image Interpretation, Computer-AssistedABSTRACT
During laparoscopic surgery, surgical gauze is usually inserted into the body cavity to help achieve hemostasis. Retention of surgical gauze in the body cavity may necessitate reoperation and increase surgical risk. Using deep learning technology, this study aimed to propose a neural network model for gauze detection from the surgical video to record the presence of the gauze. The model was trained by the training group using YOLO (You Only Look Once)v5x6, then applied to the testing group. Positive predicted value (PPV), sensitivity, and mean average precision (mAP) were calculated. Furthermore, a timeline of gauze presence in the video was drawn by the model as well as human annotation to evaluate the accuracy. After the model was well-trained, the PPV, sensitivity, and mAP in the testing group were 0.920, 0.828, and 0.881, respectively. The inference time was 11.3 ms per image. The average accuracy of the model adding a marking and filtering process was 0.899. In conclusion, surgical gauze can be successfully detected using deep learning in the surgical video. Our model provided a fast detection of surgical gauze, allowing further real-time gauze tracing in laparoscopic surgery that may help surgeons recall the location of the missing gauze.
Subject(s)
Laparoscopy , Neural Networks, Computer , Humans , Bandages , ReoperationABSTRACT
Chest X-ray (CXR) imaging is one of the most common diagnostic imaging techniques in clinical diagnosis and is usually used for radiological examinations to screen for thorax diseases. In this paper, we propose a novel computer-aided diagnosis (CAD) system based on a hybrid deep learning model composed of a convolutional neural network (CNN) and a graph neural network (GNN). The system is intended to explore implicit correlations between thorax diseases to aid in the multilabel chest X-ray image classification task, which we term â¶CheXGATâ¶. Specifically, the proposed CheXGAT framework comprises two main modules: an image representation learning (IRL) module and a graph representation learning (GRL) module. We employ the IRL module to learn high-level visual representation features from the CXR image. From the GRL module, the self-attention mechanism aggregates neighborhood features from the graphic structure to enhance the implicit correlation between thorax diseases. We adopted a data-driven method to create a disease correlation matrix that works on the message passing and aggregation process for the nodes in the GRL module. After end-to-end training, the GRL module enhances the correlation between thorax diseases to improve diagnosis performance. We performed experiments on the NIH Chest X-ray14 dataset, which contains 112,120 frontal-view radiographs; each image has multiple thorax disease labels. In the experimental results, the average AUC score of our proposed CheXGAT model reached 0.8266, and the AUC scores of Emphysema and Hernia reached 0.9447 and 0.9313, respectively. In addition, we visualized explanations via a gradient-based localization method in our proposed CheXGAT framework. Compared with previous studies, the experimental results show the competitive performance of our framework. Specifically, we propose a CAD system that uses a hybrid model to help radiologists identify CXR images from 14 chest diseases for clinical diagnosis.
Subject(s)
Deep Learning , Diagnosis, Computer-Assisted/methods , Neural Networks, Computer , Thorax/diagnostic imaging , X-RaysABSTRACT
BACKGROUND AND AIMS: The primary goals of this study were to clarify 1) the effect of weight loss by lifestyle intervention on circulating total angiopoietin-like protein 8 (ANGPTL8), and 2) the role of physical activity on serum total ANGPTL8 in northern Americans with obesity but without diabetes. METHODS AND RESULTS: A total of 130 subjects with body mass index (BMI) ⧠35 kg/m2 but without diabetes were recruited, and 121 subjects completed a weight loss program for data analysis. Abdominal adipose tissue was determined by non-contrast computed tomography (CT). Serum total ANGPTL8 was higher in the group with obesity than in the lean control group. Serum total ANGPTL8 was positively correlated with waist circumference (WC), BMI, fasting insulin, HOMA-IR, HOMA-B, QUICKI, hs-CRP, IL-6, and leptin. Serum total ANGPTL8 did not significantly differ between the two intervention groups at baseline, and it was significantly lower after weight loss, with comparable changes with diet only and diet plus physical activity. CONCLUSION: Among northern Americans with obesity but without diabetes, a lifestyle modification resulted in significant reduction of circulating total ANGPTL8 concentrations in a 6-month weight-loss period. Although addition of physical activity resulted in greater total and liver fat loss, it did not promote further significant decline of serum total ANGPTL8 beyond diet alone.
Subject(s)
Peptide Hormones , Weight Reduction Programs , Angiopoietin-Like Protein 8 , Angiopoietin-like Proteins , Body Mass Index , Exercise , Humans , Obesity/diagnosis , Obesity/therapy , Prospective Studies , Weight LossABSTRACT
BACKGROUND AND OBJECTIVE: Lung cancer is the most common cause of cancer-related death in the world. Low-dose computed tomography (LDCT) is a widely used modality in lung cancer detection. The nodule is an abnormal tissue and may evolve into lung cancer. Hence, it is crucial to detect nodules in the early detection stage. However, reviewing the LDCT scans to observe suspicious nodules is a time-consuming task. Recently, designing a computer-aided detection (CADe) system with convolutional neural network (CNN) architecture has been proven that it is helpful for radiologists. Hence, in this study, a 3-D YOLO-based CADe system, 3-D OSAF-YOLOv3, is proposed for nodule detection in LDCT images. METHODS: The proposed CADe system consists of data preprocessing, nodule detection, and non-maximum suppression algorithm (NMS). At first, the data preprocessing including the background elimination, the spacing normalization, and the volume of interest (VOI) extraction, are conducted to remove the non-lung region, normalize the image spacing, and divide LDCT image into numerous VOIs. Then, the VOIs are fed into the 3-D OSAF-YOLOv3 model, to detect the suspicious nodules. The proposed model is constructed by integrating the 3-D YOLOv3 with the one-shot aggregation module (OSA), the receptive field block (RFB), and the feature fusion scheme (FFS). Finally, the NMS algorithm is performed to eliminate the duplicated detection generated by the model. RESULTS: In this study, the LUNA-16 dataset composed 1186 nodules from 888 LDCT scans and the competition performance metric (CPM) are used to evaluate our CADe system. In the experiment results, the proposed system can achieve a sensitivities rate of 0.962 with the false positive rate of 8 and complete a CPM value of 0.905. Moreover, according to the ablation study results, the employment of OSA module, RFB, and FFS could improve the detection performance actually. Furthermore, compared to other start-of-the-art (SOTA) models, our detection system could also achieve the higher performance. CONCLUSIONS: In this study, a YOLO-based CADe system for nodule detection in CT image system integrating additional modules and scheme is proposed for nodule detection in LDCT. The result indicates that the proposed the modification can significantly improve detection performance.
Subject(s)
Lung Neoplasms , Solitary Pulmonary Nodule , Humans , Lung Neoplasms/diagnostic imaging , Radiographic Image Interpretation, Computer-Assisted/methods , Solitary Pulmonary Nodule/diagnostic imaging , Tomography, X-Ray Computed/methodsABSTRACT
OBJECTIVES: To examine the role of ADC threshold on agreement across observers and deep learning models (DLMs) plus segmentation performance of DLMs for acute ischemic stroke (AIS). METHODS: Twelve DLMs, which were trained on DWI-ADC-ADC combination from 76 patients with AIS using 6 different ADC thresholds with ground truth manually contoured by 2 observers, were tested by additional 67 patients in the same hospital and another 78 patients in another hospital. Agreement between observers and DLMs were evaluated by Bland-Altman plot and intraclass correlation coefficient (ICC). The similarity between ground truth (GT) defined by observers and between automatic segmentation performed by DLMs was evaluated by Dice similarity coefficient (DSC). Group comparison was performed using the Mann-Whitney U test. The relationship between the DSC and ADC threshold as well as AIS lesion size was evaluated by linear regression analysis. A p < .05 was considered statistically significant. RESULTS: Excellent interobserver agreement and intraobserver repeatability in the manual segmentation (all ICC > 0.98, p < .001) were achieved. The 95% limit of agreement was reduced from 11.23 cm2 for GT on DWI to 0.59 cm2 for prediction at an ADC threshold of 0.6 × 10-3 mm2/s combined with DWI. The segmentation performance of DLMs was improved with an overall DSC from 0.738 ± 0.214 on DWI to 0.971 ± 0.021 on an ADC threshold of 0.6 × 10-3 mm2/s combined with DWI. CONCLUSIONS: Combining an ADC threshold of 0.6 × 10-3 mm2/s with DWI reduces interobserver and inter-DLM difference and achieves best segmentation performance of AIS lesions using DLMs. KEY POINTS: ⢠Higher Dice similarity coefficient (DSC) in predicting acute ischemic stroke lesions was achieved by ADC thresholds combined with DWI than by DWI alone (all p < .05). ⢠DSC had a negative association with the ADC threshold in most sizes, both hospitals, and both observers (most p < .05) and a positive association with the stroke size in all ADC thresholds, both hospitals, and both observers (all p < .001). ⢠An ADC threshold of 0.6 × 10-3 mm2/s eliminated the difference of DSC at any stroke size between observers or between hospitals (p = .07 to > .99).
Subject(s)
Deep Learning , Ischemic Stroke , Stroke , Diffusion Magnetic Resonance Imaging , Humans , Ischemic Stroke/diagnostic imaging , Observer Variation , Stroke/diagnostic imagingABSTRACT
BACKGROUND: Quality indicators should be assessed and monitored to improve colonoscopy quality in clinical practice. Endoscopists must enter relevant information in the endoscopy reporting system to facilitate data collection, which may be inaccurate. The current study aimed to develop a full deep learning-based algorithm to identify and analyze intra-procedural colonoscopy quality indicators based on endoscopy images obtained during the procedure. METHODS: A deep learning system for classifying colonoscopy images for quality assurance purposes was developed and its performance was assessed with an independent dataset. The system was utilized to analyze captured images and results were compared with those of real-world reports. RESULTS: In total, 10,417 images from the hospital endoscopy database and 3157 from Hyper-Kvasir open dataset were utilized to develop the quality assurance algorithm. The overall accuracy of the algorithm was 96.72% and that of the independent test dataset was 94.71%. Moreover, 761 real-world reports and colonoscopy images were analyzed. The accuracy of electronic reports about cecal intubation rate was 99.34% and that of the algorithm was 98.95%. The agreement rate for the assessment of polypectomy rates using the electronic reports and the algorithm was 0.87 (95% confidence interval 0.83-0.90). A good correlation was found between the withdrawal time calculated using the algorithm and that entered by the physician (correlation coefficient r = 0.959, p < 0.0001). CONCLUSION: We proposed a novel deep learning-based algorithm that used colonoscopy images for quality assurance purposes. This model can be used to automatically assess intra-procedural colonoscopy quality indicators in clinical practice.
Subject(s)
Colonoscopy , Deep Learning , Algorithms , Cecum , Colonoscopy/methods , Databases, Factual , HumansABSTRACT
Lymph node metastasis also called nodal metastasis (Nmet), is a clinically primary task for physicians. The survival and recurrence of lung cancer are related to the Nmet staging from Tumor-Node-Metastasis (TNM) reports. Furthermore, preoperative Nmet prediction is still a challenge for the patient in managing the surgical plan and making treatment decisions. We proposed a multi-energy level fusion model with a principal feature enhancement (PFE) block incorporating radiologist and computer science knowledge for Nmet prediction. The proposed model is custom-designed by gemstone spectral imaging (GSI) with different energy levels on dual-energy computer tomography (CT) from a primary tumor of lung cancer. In the experiment, we take three different energy level fusion datasets: lower energy level fusion (40, 50, 60, 70 keV), higher energy level fusion (110, 120, 130, 140 keV), and average energy level fusion (40, 70, 100, 140 keV). The proposed model is trained by lower energy level fusion that is 93% accurate and the value of Kappa is 86%. When we used the lower energy level images to train the fusion model, there has been a significant difference to other energy level fusion models. Hence, we apply 5-fold cross-validation, which is used to validate the performance result of the multi-keV model with different fusion datasets of energy level images in the pathology report. The cross-validation result also demonstrates that the model with the lower energy level dataset is more robust and suitable in predicting the Nmet of the primary tumor. The lower energy level shows more information of tumor angiogenesis or heterogeneity provided the proposed fusion model with a PFE block and channel attention blocks to predict Nmet from primary tumors.
Subject(s)
Deep Learning , Lung Neoplasms , Computers , Humans , Lung Neoplasms/diagnostic imaging , Lymphatic Metastasis/diagnostic imaging , Tomography, X-Ray Computed/methodsABSTRACT
OBJECTIVES: Visualization and photodocumentation during endoscopy procedures are suggested to be one indicator for endoscopy performance quality. However, this indicator is difficult to measure and audit manually in clinical practice. Artificial intelligence (AI) is an emerging technology that may solve this problem. METHODS: A deep learning model with an accuracy of 96.64% was developed from 15,305 images for upper endoscopy anatomy classification in the unit. Endoscopy images for asymptomatic patients receiving screening endoscopy were evaluated with this model to assess the completeness of photodocumentation rate. RESULTS: A total of 15,723 images from 472 upper endoscopies performed by 12 endoscopists were enrolled. The complete photodocumentation rate from the pharynx to the duodenum was 53.8% and from the esophagus to the duodenum was 78.0% in this study. Endoscopists with a higher adenoma detection rate had a higher complete examination rate from the pharynx to duodenum (60.0% vs. 38.7%, P < 0.0001) and from esophagus to duodenum (83.0% vs. 65.7%, P < 0.0001) compared with endoscopists with lower adenoma detection rate. The pharynx, gastric angle, gastric retroflex view, gastric antrum, and the first portion of duodenum are likely to be missed by endoscopists with lower adenoma detection rates. CONCLUSIONS: We report the use of a deep learning model to audit endoscopy photodocumentation quality in our unit. Endoscopists with better performance in colonoscopy had a better performance for this quality indicator. The use of such an AI system may help the endoscopy unit audit endoscopy performance.
Subject(s)
Adenoma , Deep Learning , Adenoma/diagnosis , Artificial Intelligence , Colonoscopy/methods , Endoscopy, Gastrointestinal , HumansABSTRACT
BACKGROUND: Photodocumentation during endoscopy procedures is one of the indicators for endoscopy performance quality; however, this indicator is difficult to measure and audit in the endoscopy unit. Emerging artificial intelligence technology may solve this problem, which requires a large amount of material for model development. We developed a deep learning-based endoscopic anatomy classification system through convolutional neural networks with an accelerated data preparation approach. PATIENTS AND METHODS: We retrospectively collected 8,041 images from esophagogastroduodenoscopy (EGD) procedures and labeled them using two experts for nine anatomical locations of the upper gastrointestinal tract. A base model for EGD image multiclass classification was first developed, and an additional 6,091 images were enrolled and classified by the base model. A total of 5,963 images were manually confirmed and added to develop the subsequent enhanced model. Additional internal and external endoscopy image datasets were used to test the model performance. RESULTS: The base model achieved total accuracy of 96.29%. For the enhanced model, the total accuracy was 96.64%. The overall accuracy improved with the enhanced model compared with the base model for the internal test dataset without narrowband images (93.05% vs. 91.25%, p < 0.01) or with narrowband images (92.74% vs. 90.46%, p < 0.01). The total accuracy was 92.56% of the enhanced model on the external test dataset. CONCLUSIONS: We constructed a deep learning-based model with an accelerated approach that can be used for quality control in endoscopy units. The model was also validated with both internal and external datasets with high accuracy.
Subject(s)
Artificial Intelligence , Deep Learning , Endoscopy, Gastrointestinal/methods , Humans , Neural Networks, Computer , Retrospective StudiesABSTRACT
In this study, the performance of machine learning in classifying parotid gland tumors based on diffusion-related features obtained from the parotid gland tumor, the peritumor parotid gland, and the contralateral parotid gland was evaluated. Seventy-eight patients participated in this study and underwent magnetic resonance diffusion-weighted imaging. Three regions of interest, including the parotid gland tumor, the peritumor parotid gland, and the contralateral parotid gland, were manually contoured for 92 tumors, including 20 malignant tumors (MTs), 42 Warthin tumors (WTs), and 30 pleomorphic adenomas (PMAs). We recorded multiple apparent diffusion coefficient (ADC) features and applied a machine-learning method with the features to classify the three types of tumors. With only mean ADC of tumors, the area under the curve of the classification model was 0.63, 0.85, and 0.87 for MTs, WTs, and PMAs, respectively. The performance metrics were improved to 0.81, 0.89, and 0.92, respectively, with multiple features. Apart from the ADC features of parotid gland tumor, the features of the peritumor and contralateral parotid glands proved advantageous for tumor classification. Combining machine learning and multiple features provides excellent discrimination of tumor types and can be a practical tool in the clinical diagnosis of parotid gland tumors.