Diagnostic value of a computer-assisted diagnosis system for the ultrasound features in thyroid nodules

ABSTRACT Objective: To explore the diagnostic value of the TUIAS (SW_TH01/II) computer-aided diagnosis (CAD) software system for the ultrasound Thyroid Imaging Reporting and Data System (TI-RADS) features in thyroid nodules. Materials and methods: This retrospective study enrolled patients with thyroid nodules in Shanghai East Hospital between January 2017 and October 2021. The novel CAD software (SW_TH01/II) and three sonographers performed a qualitative analysis of the ultrasound TI-RADS features in aspect ratio, margin irregularity, margin smoothness, calcification, and echogenicity of the thyroid nodules. Results: A total of 225 patients were enrolled. The accuracy, sensitivity, and specificity of the CAD software in "aspect ratio" were 95.6%, 96.2%, and 95.4%, in "margin irregularity" were 90.7%, 90.5%, and 90.9%, in "margin smoothness" were 85.8%, 88.5%, and 83.0%, in "calcification" were 83.6%, 81.7%, and 82.0%, in "homogeneity" were 88.9%, 90.6%, and 82.2%, in "major echo" were 85.3%, 88.0%, and 85.4%, and in "contains very hypoechoic echo" were 92.0%, 90.0%, and 92.4%. The analysis time of the CAD software was significantly shorter than for the sonographers (2.7 ± 1.6 vs. 29.7 ± 12.7 s, P < 0.001). Conclusion: The CAD system achieved high accuracy in describing thyroid nodule features. It might assist in clinical thyroid nodule analysis.

Application progress of machine learning in orthopedic diagnosis and treatment / 国际生物医学工程杂志

In recent years, artificial intelligence-related technologies have been deeply combined with many medical fields, and the intersection of medicine and engineering has become a hot topic. There are problems with heavy data and difficulty making decisions in orthopedic disease diagnosis and treatment. Machine learning is an important method of artificial intelligence. Since it can automatically analyze and predict medical big data, it is widely used in the field of orthopedics. It also assists physicians in completing disease diagnosis and treatment efficiently. In this review paper, the application and progress of machine learning in preoperative, intraoperative, and postoperative diagnosis and treatment in orthopedics are reviewed, providing new ways for exploring more rational diagnosis and treatment strategies.

Research progress on medical image dataset expansion methods / 生物医学工程学杂志

Computer-aided diagnosis (CAD) systems play a very important role in modern medical diagnosis and treatment systems, but their performance is limited by training samples. However, the training samples are affected by factors such as imaging cost, labeling cost and involving patient privacy, resulting in insufficient diversity of training images and difficulty in data obtaining. Therefore, how to efficiently and cost-effectively augment existing medical image datasets has become a research hotspot. In this paper, the research progress on medical image dataset expansion methods is reviewed based on relevant literatures at home and abroad. First, the expansion methods based on geometric transformation and generative adversarial networks are compared and analyzed, and then improvement of the augmentation methods based on generative adversarial networks are emphasized. Finally, some urgent problems in the field of medical image dataset expansion are discussed and the future development trend is prospected.

Combination of S-Detect and strain elastography in the conventional ultrasonic diagnosis of breast tumors / 中国综合临床

Objective:To explore the value of s-detect combined with elastography in the differential diagnosis of benign and malignant breast tumors.Methods:The ultrasound diagnosis data of 136 patients with breast mass examined by ultrasound in the First Affiliated Hospital of Shantou University Medical College from April 2018 to January 2020 were retrospectively analyzed. The breast lesions diagnosed as BI-RADS 3 or above were analyzed by conventional ultrasound, strain elastic imaging strain ratio (SR) and S-Detect Computer-aided Diagnosis (CAD) technology successively and were used for cross-sectional study. The corresponding benign and malignant judgment results were obtained, and the efficacy of individual diagnosis and combined diagnosis were compared and analyzed.Results:Conventional ultrasound, SR, S-Detect alone and conventional ultrasound+SR, conventional ultrasound+S-detect, conventional ultrasound + S-detect +SR combined diagnosis of breast tumors, the area under the receiver operating curve (AREA under the receiver operating curve) Characteristic curve, AUC) were 0.776, 0.839, 0.802, 0.861, 0.832 and 0.870, respectively. SR, S-Detect, conventional ultrasound +SR, conventional ultrasound + S-detect and conventional ultrasound +SR+ S-detect were compared with conventional ultrasound group, Z values were 1.49, 0.70, 2.76, 2.52, 2.96, respectively, and P values were 0.137, 0.484, 0.006, 0.012 and 0.003, respectively. The difference was statistically significant. The accuracy of conventional ultrasound +S-Detect+SR was the highest (84.1%), compared with pathological results, its Kappa value was 0.687, showing the strongest consistency. Conclusion:S-detect combined with strain elastography assisted by conventional ultrasound can significantly improve the diagnostic efficiency of benign and malignant breast tumors.

Ultrasonic automatic classification algorithm in the diagnosis of breast masses / 中华超声影像学杂志

Objective:To explore the clinical effectiveness of an automatic computer-aided diagnosis(CAD) method in benign and malignant breast masses discrimination.Methods:The ultrasound images of 539 patients from the Second Hospital of Harbin Medical University and the Second Hospital of Hebei Medical University between 2012 to 2019 were analyzed retrospectively. According to the fifth Breast Imaging Reporting and Data System (BI-RADS), four breast radiologists first sent the case into a BI-RADS category with the original ultrasound image. Then with the CAD result, radiologists gave a category again. Pathology results and clinical data were not available to the radiologists during the diagnosis process. The histological and follow-up results were used as the gold standard. The accuracy of CAD automatic classification, radiologists′ diagnosis before and after CAD application were compared using the ROC curves. The accuracy, sensitivity and specificity of the diagnosis were also calculated.Results:The classification algorithm has a good performance in benign and malignant breast masses discrimination.When the cutoff value was 0.495, the accuracy, sensitivity and specificity were 0.878, 0.868 and 0.886 respectively. When the cutoff value was 0.203, the sensitivity and specificity was 0.981 and 0.337 respectively. With the CAD method, the radiologists improved their diagnostic performance. The total area under the ROC curve for the four radiologists increased from 0.775 to 0.871( P<0.001). The total sensitivity increased from 0.786 to 0.842, and the specificity increased from 0.681 to 0.813. Conclusions:The automatic classification algorithm in this study provides quantitative reference for doctors′ diagnosis. It has the potential to improve junior radiologists′ diagnostic performance in differentiating benign and malignant breast masses.

Artificial intelligence-assisted colonoscopy: a narrative review of current data and clinical applications

Colonoscopy is the reference standard procedure for the prevention and diagnosis of colorectal cancer, which is a leading cause of cancer-related deaths in Singapore. Artificial intelligence systems are automated, objective and reproducible. Artificial intelligence-assisted colonoscopy has recently been introduced into clinical practice as a clinical decision support tool. This review article provides a summary of the current published data and discusses ongoing research and current clinical applications of artificial intelligence-assisted colonoscopy.

Computer-aided intelligent diagnosis and education system for cardiac magnetic resonance imaging / 中华医学教育探索杂志

Cardiac magnetic resonance imaging (CMRI) can comprehensively observe the anatomical structure, motion function and tissue characteristics of the heart through multi-parameter and multi-plane sequence imaging, which has become the gold standard for the diagnosis of cardiomyopathy. However, there is a serious shortage of high-level imaging doctors who can diagnose CMRI due to the numerous sequences and difficult fusion of CMRI. To realize the personalized teaching and interactive education, we have designed a platform that could use artificial intelligence to pre-process medical images for clinical diagnosis, and localize the week points of knowledge, thus improving the teaching effect of cardiac imaging diagnosis.

Progress in computer-assisted Alberta stroke program early computer tomography score of acute ischemic stroke based on different modal images / 生物医学工程学杂志

Clinically, non-contrastive computed tomography (NCCT) is used to quickly diagnose the type and area of ​​stroke, and the Alberta stroke program early computer tomography score (ASPECTS) is used to guide the next treatment. However, in the early stage of acute ischemic stroke (AIS), it's difficult to distinguish the mild cerebral infarction on NCCT with the naked eye, and there is no obvious boundary between brain regions, which makes clinical ASPECTS difficult to conduct. The method based on machine learning and deep learning can help physicians quickly and accurately identify cerebral infarction areas, segment brain areas, and operate ASPECTS quantitative scoring, which is of great significance for improving the inconsistency in clinical ASPECTS. This article describes current challenges in the field of AIS ASPECTS, and then summarizes the application of computer-aided technology in ASPECTS from two aspects including machine learning and deep learning. Finally, this article summarizes and prospects the research direction of AIS-assisted assessment, and proposes that the computer-aided system based on multi-modal images is of great value to improve the comprehensiveness and accuracy of AIS assessment, which has the potential to open up a new research field for AIS-assisted assessment.

The latest research progress on early diagnosis of lung cancer according to CT-based computer intelligent analysis / 中国胸心血管外科临床杂志

@#Lung cancer has brought tough challenges to human health due to its high incidence and mortality rate in the current practice. Nowadays, computed tomography (CT) imaging is still the most preferred diagnostic tool for early screening of lung cancer. However, a great challenge brought from accumulative CT imaging data can not meet the demand of the current clinical practice. As a novel kind of artificial intelligence technique aimed to deal with medical images, a computer-aided diagnosis has been found to provide useful auxiliary information, attenuate the workload of doctors, and significantly improve the efficiency and accuracy for clinical diagnosis of lung cancer. Therefore, an effective combination of computer-aided techniques and CT imaging has increasingly become an active area of investigation in early diagnosis of lung cancer. This review aims to summarize the latest progress on the diagnostic value of computer-aided technology with regard to early stage lung cancer from the perspectives of machine learning and deep learning.

External and Internal Validation of a Computer Assisted Diagnostic Model for Detecting Multi-Organ Mass Lesions in CT images / 中国医学科学杂志(英文版)

Objective We developed a universal lesion detector (ULDor) which showed good performance in in-lab experiments. The study aims to evaluate the performance and its ability to generalize in clinical setting via both external and internal validation. Methods The ULDor system consists of a convolutional neural network (CNN) trained on around 80K lesion annotations from about 12K CT studies in the DeepLesion dataset and 5 other public organ-specific datasets. During the validation process, the test sets include two parts: the external validation dataset which was comprised of 164 sets of non-contrasted chest and upper abdomen CT scans from a comprehensive hospital, and the internal validation dataset which was comprised of 187 sets of low-dose helical CT scans from the National Lung Screening Trial (NLST). We ran the model on the two test sets to output lesion detection. Three board-certified radiologists read the CT scans and verified the detection results of ULDor. We used positive predictive value (PPV) and sensitivity to evaluate the performance of the model in detecting space-occupying lesions at all extra-pulmonary organs visualized on CT images, including liver, kidney, pancreas, adrenal, spleen, esophagus, thyroid, lymph nodes, body wall, thoracic spine,

Research on computer aided diagnosis of otitis media based on faster region convolutional neural network / 生物医学工程学杂志

Otitis media is one of the common ear diseases, and its accurate diagnosis can prevent the deterioration of conductive hearing loss and avoid the overuse of antibiotics. At present, the diagnosis of otitis media mainly relies on the doctor's visual inspection based on the images fed back by the otoscope equipment. Due to the quality of otoscope equipment pictures and the doctor's diagnosis experience, this subjective examination has a relatively high rate of misdiagnosis. In response to this problem, this paper proposes the use of faster region convolutional neural networks to analyze clinically collected digital otoscope pictures. First, through image data enhancement and preprocessing, the number of samples in the clinical otoscope dataset was expanded. Then, according to the characteristics of the otoscope picture, the convolutional neural network was selected for feature extraction, and the feature pyramid network was added for multi-scale feature extraction to enhance the detection ability. Finally, a faster region convolutional neural network with anchor size optimization and hyperparameter adjustment was used for identification, and the effectiveness of the method was tested through a randomly selected test set. The results showed that the overall recognition accuracy of otoscope pictures in the test samples reached 91.43%. The above studies show that the proposed method effectively improves the accuracy of otoscope picture classification, and is expected to assist clinical diagnosis.

Deep-learning based computer aided diagnosis system in detecting fractures on anteroposterior chest DR films / 中国介入影像与治疗学

Objective: To evaluate the efficiency of deep-learning based computer aided diagnosis system (DL-CAD) in detecting fractures on DR chest anteroposterior films, and to explore its capability of assisting the junior radiologist. Methods: ①Experiment 1: A total of 547 DR chest anteroposterior films, including 361 patients with 983 chest fractures and 186 without chest fractures were retrospectively analyzed. The predictive performance of DL-CAD for fracture was evaluated. ②Experiment 2: Totally 397 patients were randomly selected from experiment 1, including 211 cases with 604 chest fractures and 186 cases without chest fractures. The results of DL-CAD alone (group 1), a junior radiology resident alone (group 2), a junior radiology resident aided with DL-CAD (group 3) and a senior radiologist alone (group 4) were recorded and compared, respectively. Results: ①For experiment 1: Among 983 fractures, DL-CAD identified 672 fractures, 641 were correctly identified and 31 were misdiagnosed, with a sensitivity of 65.21% (641/983) and F-measure of 77.46%. Out of a total of 361 fracture cases, DL-CAD identified 314 cases, misdiagnosed 6 cases, with a sensitivity of 86.98% (314/361) and F-measure of 92.22%. ②Experiment 2: The sensitivity of fracture detection was 62.09% (375/604), 61.59% (372/604), 86.75% (524/604) and 83.44% (504/604), and the F-measure was 75.38%, 74.62%, 90.74%, 89.84% for group 1, 2, 3 and 4, respectively. The detection efficacy of group 3 and 4 were both higher than that of group 1 and 2 (all P0.05). Conclusion: DL-CAD software showed good detection effect of fractures on DR chest anteroposterior films, which could effectively improve the diagnostic performance of junior radiologist in fracture detection.

Application of AmCAD-UT Detection in ultrasonic diagnosis of thyroid nodules / 中国医学影像技术

Objective: To evaluate the diagnostic efficiency and clinical application value of computer-aided diagnosis (CAD) system AmCAD-UT Detection in thyroid ultrasound examination. Methods: Totally 171 patients with thyroid nodules requiring ultrasonic examination were collected. Ultrasonic thyroid images of all patients were obtained, then were analyzed with AmCAD-UT Detection only, 4 ultrasound physicians (A, B, C, D, having more than 10 years, 5 years, 1 year and 1 month experience, respectively) with or without AmCAD-UT Detection, respectively, and the nodules were classified according to the American College of Radiology Thyroid Imaging Reporting and Data System guidelines (ACR-TIRADS).Taken pathologic results as the gold standards, ROC curves of the classification of nodules of mCAD-UT Detection and 4 radiologists using the former or not were drawn according to ACR-TIRADS, and the optimal cut-off value for diagnosis of nodule malignancy with ACR-TIRADS guidelines and AUC were calculated, and their diagnostic efficacy were then analyzed. Results: A total of 205 thyroid nodules were involved, with 89 benign and 116 malignant lesions. TR5 was the optimal cut-off value for diagnosis of benign or malignant nodule with ACR-TIRADS. The diagnostic sensitivity of AmCAD-UT Detection to diagnose thyroid malignant nodule was similar to that of physician B (P=1.00), and the specificity was lower than that of physician A and B (both P<0.05), while its AUC was statistically different with physician A, B and D (Z=4.34, 3.71, 2.76, all P<0.05). With AmCAD-UT Detection, the sensitivity (93.10%, 90.52%, 85.34%, 75.00%) and AUC values (0.95, 0.93, 0.86, 0.86) of diagnosis of thyroid nodules of all physicians were improved (all P<0.05), while for the specificity, only physician C and D had better results (both P<0.05). Conclusion: Thyroid CAD system AmCAD-UT Detection has certain value for diagnosing thyroid nodules, with sensitivity similar to physician with 5 years ultrasound diagnostic experience, therefore can be used to improve diagnosis efficiency of thyroid nodule of physicians, especially for those with less experiences.

Computer-aided quantitative evaluation of texture features extracted from two-dimensional high frequency ultrasonograms after acute muscle contusion / 第二军医大学学报

Objective: To quantitatively evaluate the changes of texture features extracted from two-dimensional high frequency ultrasonograms of human muscle injured by acute contusion using the multiscale decomposition of echo intensity of interface reflections, and to preliminarily explore its clinical value. Methods: Two-dimensional ultrasound images of local muscles of 10 male patients with acute upper limb muscle contusion were obtained using high-frequency ultrasound. The region of interest (ROI) of normal muscle texture and the ROI of muscle texture with suspected injury on the same image of the patients were selected by Matlab 7.0 software in offline state. Eight texture parameters including mean of gray scale (Mean), standard variance of gray scale (SDev), number of blobs (NOB) of texture density, irregularity (IRGL) of texture primitive shape, mean size of blobs (SOB) of texture primitive, homogeneity of distribution (HOD) of texture uniformity, directionality of texture distribution (DOD) and periodicity of texture distribution (POD) of the two ROIs were extracted. The similarity difference values of the eight texture parameters between the two ROIs were automatically calculated by the multiscale decomposition of echo intensity of interface reflections. Two-dimensional ultrasound images of normal muscles in the same part of 10 healthy male volunteers were selected as controls, and two ROIs were randomly selected to calculate the similarity difference values of the above eight texture features between them. The similarity difference values of the eight texture features between patients with upper limb muscle contusion and healthy volunteers were compared. Results: Local hyperechoic lesions were found with disordered muscle fibers and fuzzy textures in the patients with acute upper extremity muscle contusion. There were significant differences in the similarity difference of fve textural parameters (IRGL, DOD, POD, Mean and SDev) between patients with acute upper limb muscle contusion and healthy controls (P<0.01). Conclusion: Computer-aided quantitative evaluation based on multiscale decomposition of echo intensity of interface reflections can lead to more accurate and detailed quantitative diagnosis of texture features extracted from two-dimensional high frequency ultrasonic images of muscle injured by acute contusion than human eyes, and it may have clinical values.

Coupled convolutional and graph network-based diagnosis of Alzheimer's disease using MRI / 南方医科大学学报

OBJECTIVE@#To propose a coupled convolutional and graph convolutional network (CCGCN) model for diagnosis of Alzheimer's disease (AD) and its prodromal stage.@*METHODS@#The disease-related brain regions generated by group-wise comparison were used as the input. The convolutional neural networks (CNNs) were used to extract disease-related features from different locations on brain magnetic resonance (MR) images. The generated features via the graph convolutional network (GCN) were processed, and graph pooling was performed to analyze the inherent relationship between the brain topology and the diagnosis task adaptively. Through ADNI dataset, we acquired the accuracy, sensitivity and specificity of the diagnosis tasks for AD and its prodromal stages, followed by an ablation study on the model structure.@*RESULTS@#The CCGCN model outperformed the current state-of-the-art methods and showed a classification accuracy of 92.5% for AD with a sensitivity of 88.1% and a specificity of 96.0%.@*CONCLUSIONS@#Based on the structural and topological features of the brain MR images, the proposed CCGCN model shows excellent performance in AD diagnosis and is expected to provide important assistance to physicians in disease diagnosis.

Applications of Artificial Intelligence in Musculoskeletal System Imaging / 中国医学科学院学报

Artificial intelligence (AI) represents the latest wave of computer revolution and is considered revolutionary technology in many industries including healthcare. AI has been applied in medical imaging mainly due to the improvement of computational learning,big data mining,and innovations of neural network architecture. AI can improve the efficiency and accuracy of imaging diagnosis and reduce medical cost;also,it can be used to predict the disease risk. In this article we summarize and analyze the application of AI in musculoskeletal imaging.

Computer-Aided Diagnosis System for the Evaluation of Thyroid Nodules on Ultrasonography: Prospective Non-Inferiority Study according to the Experience Level of Radiologists

OBJECTIVE: To determine whether a computer-aided diagnosis (CAD) system for the evaluation of thyroid nodules is non-inferior to radiologists with different levels of experience.MATERIALS AND METHODS: Patients with thyroid nodules with a decisive diagnosis of benign or malignant nodule were consecutively enrolled from November 2017 to September 2018. Three radiologists with different levels of experience (1 month, 4 years, and 7 years) in thyroid ultrasound (US) reviewed the thyroid US with and without using the CAD system. Statistical analyses included non-inferiority testing of the diagnostic accuracy for malignant thyroid nodules between the CAD system and the three radiologists with a non-inferiority margin of 10%, comparison of the diagnostic performance, and the added value of the CAD system to the radiologists.RESULTS: Altogether, 197 patients were included in the study cohort. The diagnostic accuracy of the CAD system (88.48%, 95% confidence interval [CI] = 82.65–92.53) was non-inferior to that of the radiologists with less experience (1 month and 4 year) of thyroid US (83.03%, 95% CI = 76.52–88.02; p < 0.001), whereas it was inferior to that of the experienced radiologist (7 years) (95.76%, 95% CI = 91.37–97.96; p = 0.138). The sensitivity and negative predictive value of the CAD system were significantly higher than those of the less-experienced radiologists were, whereas no significant difference was found with those of the experienced radiologist. A combination of US and the CAD system significantly improved sensitivity and negative predictive value, although the specificity and positive predictive value deteriorated for the less-experienced radiologists.CONCLUSION: The CAD system may offer support for decision-making in the diagnosis of malignant thyroid nodules for operators who have less experience with thyroid US.

Web-Based Spine Segmentation Using Deep Learning in Computed Tomography Images / 대한의료정보학회지

Study on classification and identification of depressed patients and healthy people among adolescents based on optimization of brain characteristics of network / 生物医学工程学杂志

To enhance the accuracy of computer-aided diagnosis of adolescent depression based on electroencephalogram signals, this study collected signals of 32 female adolescents (16 depressed and 16 healthy, age: 16.3 ± 1.3) with eyes colsed for 4 min in a resting state. First, based on the phase synchronization between the signals, the phase-locked value (PLV) method was used to calculate brain functional connectivity in the θ and α frequency bands, respectively. Then based on the graph theory method, the network parameters, such as strength of the weighted network, average characteristic path length, and average clustering coefficient, were calculated separately (

Feasibility of fully automated classification of whole slide images based on deep learning

Although microscopic analysis of tissue slides has been the basis for disease diagnosis for decades, intra- and inter-observer variabilities remain issues to be resolved. The recent introduction of digital scanners has allowed for using deep learning in the analysis of tissue images because many whole slide images (WSIs) are accessible to researchers. In the present study, we investigated the possibility of a deep learning-based, fully automated, computer-aided diagnosis system with WSIs from a stomach adenocarcinoma dataset. Three different convolutional neural network architectures were tested to determine the better architecture for tissue classifier. Each network was trained to classify small tissue patches into normal or tumor. Based on the patch-level classification, tumor probability heatmaps can be overlaid on tissue images. We observed three different tissue patterns, including clear normal, clear tumor and ambiguous cases. We suggest that longer inspection time can be assigned to ambiguous cases compared to clear normal cases, increasing the accuracy and efficiency of histopathologic diagnosis by pre-evaluating the status of the WSIs. When the classifier was tested with completely different WSI dataset, the performance was not optimal because of the different tissue preparation quality. By including a small amount of data from the new dataset for training, the performance for the new dataset was much enhanced. These results indicated that WSI dataset should include tissues prepared from many different preparation conditions to construct a generalized tissue classifier. Thus, multi-national/multi-center dataset should be built for the application of deep learning in the real world medical practice.