Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Automatic Identification of Information Quality Metrics in Health News Stories.

In Frontiers in public health

Objective: Many online and printed media publish health news of questionable trustworthiness and it may be difficult for laypersons to determine the information quality of such articles. The purpose of this work was to propose a methodology for the automatic assessment of the quality of health-related news stories using natural language processing and machine learning. Materials and Methods: We used a database from the website HealthNewsReview.org that aims to improve the public dialogue about health care. HealthNewsReview.org developed a set of criteria to critically analyze health care interventions' claims. In this work, we attempt to automate the evaluation process by identifying the indicators of those criteria using natural language processing-based machine learning on a corpus of more than 1,300 news stories. We explored features ranging from simple n-grams to more advanced linguistic features and optimized the feature selection for each task. Additionally, we experimented with the use of pre-trained natural language model BERT. Results: For some criteria, such as mention of costs, benefits, harms, and "disease-mongering," the evaluation results were promising with an F1 measure reaching 81.94%, while for others the results were less satisfactory due to the dataset size, the need of external knowledge, or the subjectivity in the evaluation process. Conclusion: These used criteria are more challenging than those addressed by previous work, and our aim was to investigate how much more difficult the machine learning task was, and how and why it varied between criteria. For some criteria, the obtained results were promising; however, automated evaluation of the other criteria may not yet replace the manual evaluation process where human experts interpret text senses and make use of external knowledge in their assessment.

Al-Jefri Majed, Evans Roger, Lee Joon, Ghezzi Pietro

2020

health information quality assessment, machine learning, natural language processing, online health information, text classification

Radiology Radiology

Prognostic Performance of Albumin-Bilirubin Grade With Artificial Intelligence for Hepatocellular Carcinoma Treated With Transarterial Chemoembolization Combined With Sorafenib.

In Frontiers in oncology

Purpose : To establish albumin-bilirubin (ALBI) grade-based and Child-Turcotte-Pugh (CTP) grade-based nomograms, as well as to develop an artificial neural network (ANN) model to compare the prognostic performance and discrimination of these two grades for hepatocellular carcinoma (HCC) treated with transarterial chemoembolization (TACE) combined with sorafenib as an initial treatment.

Methods : This multicenter retrospective study included patients from three hospitals between January 2013 and August 2018. In the training cohort, independent risk factors associated with overall survival (OS) were identified by univariate and multivariate analyses. The nomograms and ANN were established and then validated in two validation cohorts.

Results : A total of 504 patients (319, 61, and 124 patients from hospitals A, B, and C, respectively) were included. The median OS was 15.2, 26.9, and 14.8 months in the training cohort and validation cohorts 1 and 2, respectively (P = 0.218). In the training cohort, both ALBI grade and CTP grade were identified as independent risk factors. The ALBI grade-based and CTP grade-based nomograms were established separately and showed similar prognostic performance and discrimination when validated in the validation cohorts (C-index in validation cohort 1: 0.799 vs. 0.779, P = 0.762; in validation cohort 2: 0.700 vs. 0.693, P = 0.803). The ANN model showed that the ALBI grade had higher importance in survival prediction than the CTP grade.

Conclusions : The ALBI grade and CTP grade have comparable prognostic performance for HCC patients treated with TACE combined with sorafenib. ALBI grades 1 and 2 have the potential to act as a stratification factor for clinical trials on the combination therapy of TACE and systemic therapy.

Zhong Bin-Yan, Yan Zhi-Ping, Sun Jun-Hui, Zhang Lei, Hou Zhong-Heng, Yang Min-Jie, Zhou Guan-Hui, Wang Wan-Sheng, Li Zhi, Huang Peng, Zhang Shen, Zhu Xiao-Li, Ni Cai-Fang

2020

albumin–bilirubin, artificial intelligence, artificial neural network, hepatocellular carcinoma, nomogram

Radiology Radiology

Deep learning reconstruction versus iterative reconstruction for cardiac CT angiography in a stroke imaging protocol: reduced radiation dose and improved image quality.

In Quantitative imaging in medicine and surgery

Background : To assess the radiation dose and image quality of cardiac computed tomography angiography (CCTA) in an acute stroke imaging protocol using a deep learning reconstruction (DLR) method compared to a hybrid iterative reconstruction algorithm.

Methods : Retrospective analysis of 296 consecutive patients admitted to the emergency department for stroke suspicion. All patients underwent a stroke CT imaging protocol including a non-enhanced brain CT, a brain perfusion CT imaging if necessary, a CT angiography (CTA) of the supra-aortic vessels, a CCTA and a post-contrast brain CT. The CCTA was performed with a prospectively ECG-gated volume acquisition. Among all CT scans performed, 143 were reconstructed with an iterative reconstruction algorithm (AIDR 3D, adaptive iterative dose reduction three dimensional) and 146 with a DLR algorithm (AiCE, advanced intelligent clear-IQ engine). Image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and subjective image quality (IQ) scored from 1 to 4 were assessed. Dose-length product (DLP), volume CT dose index (CTDIvol) and effective dose (ED) were obtained.

Results : The radiation dose was significantly lower with AiCE than with AIDR 3D (DLP =106.4±50.0 vs. 176.1±37.1 mGy·cm, CTDIvol =6.9±3.2 vs. 11.5±2.2 mGy, and ED =1.5±0.7 vs. 2.5±0.5 mSv) (P<0.001). The median SNR and CNR were higher [9.9 (IQR, 8.1-12.3); and 12.6 (IQR, 10.5-15.5), respectively], with AiCE than with AIDR 3D [6.5 (IQR, 5.2-8.5); and 8.4 (IQR, 6.7-11.0), respectively] (P<0.001). SNR and CNR were increased by 51% and 49%, respectively, with AiCE compared to AIDR 3D. The image quality was significantly better with AiCE (mean IQ score =3.4±0.7) than with AIDR 3D (mean IQ score =3±0.9) (P<0.001).

Conclusions : The use of a DLR algorithm for cardiac CTA in an acute stroke imaging protocol reduced the radiation dose by about 40% and improved the image quality by about 50% compared to an iterative reconstruction algorithm.

Bernard Angélique, Comby Pierre-Olivier, Lemogne Brivaël, Haioun Karim, Ricolfi Frédéric, Chevallier Olivier, Loffroy Romaric

2021-Jan

Computed tomography angiography (CTA), artificial intelligence, cardiac imaging, deep learning, image reconstruction

Surgery Surgery

Evaluation of magnetic resonance image segmentation in brain low-grade gliomas using support vector machine and convolutional neural network.

In Quantitative imaging in medicine and surgery

Background : Image segmentation of brain low-grade glioma (LGG) magnetic resonance imaging (MRI) contributes tremendously to diagnosis, classification and treatment of the disease. A tangible, accurate, reliable and fast image segmentation technique is demanded in clinical diagnosis and research.

Methods : The emerging machine learning technique has been demonstrated its unique capability in the field of medical image processing, including medical image segmentation. Support vector machine (SVM) and convolutional neural network (CNN) are two widely used machine learning methods. In this work, image segmentation tools based on SVM and CNN are developed and evaluated for brain LGG MR image segmentation studies. The segmentation performance in terms of accuracy and cost is quantitatively analyzed and compared between the SVM and CNN techniques developed.

Results : Computed on the Google CoLab, each of the 109 SVM models represents an individual patient, is trained using a single image of that patient and takes a few seconds to complete. The CNN model is trained on a drastically larger dataset of 19,760 data augmented images and takes approximately 2 hours to obtain the most optimal result. The SVM models achieved an average and median accuracy of 0.937 and 0.976 respectively, precision of 0.456 and 0.535 respectively, recall of 0.878 and 0.906 respectively, and F1 score of 0.546 and 0.662 respectively. Although the CNN model required a significantly longer calculation time, it surpassed the SVM models in performance in LGG MR image segmentation, achieving an accuracy of 0.998, a precision of 0.999, a recall of 0.999 and an F1 score of 0.999.

Conclusions : This study shows that SVM with appropriate filtering techniques is capable of obtaining reliable and fast segmentation of brain LGG MR images with sufficient accuracy and limited image data. CNN technique outperforms SVM in the accuracy of segmentation with requirements of significantly enlarged data set, long computation time and high-performance computer.

Yang Qifan, Zhang Huijuan, Xia Jun, Zhang Xiaoliang

2021-Jan

Magnetic resonance imaging (MRI), convolutional neural network (CNN), image segmentation, low-grade glioma (LGG), machine learning, support vector machine (SVM)

Radiology Radiology

Fully automatic deep learning trained on limited data for carotid artery segmentation from large image volumes.

In Quantitative imaging in medicine and surgery

Background : The objectives of this study were to develop a 3D convolutional deep learning framework (CarotidNet) for fully automatic segmentation of carotid bifurcations in computed tomography angiography (CTA) images and to facilitate the quantification of carotid stenosis and risk assessment of stroke.

Methods : Our pipeline was a two-stage cascade network that included a localization phase and a segmentation phase. The network framework was based on the 3D version of U-Net, but was refined in three ways: (I) by adding residual connections and a deep supervision strategy to cope with the vanishing problem in back-propagation; (II) by adopting dilated convolution in order to strengthen the capacity to capture contextual information; and (III) by establishing a hybrid objective function to address the extreme imbalance between foreground and background voxels.

Results : We trained our networks on 15 cases and evaluated their performance based on 41 cases from the MICCAI Challenge 2009 dataset. A Dice similarity coefficient of 82.3% was achieved for the test cases.

Conclusions : We developed a carotid segmentation method based on U-Net that can segment tiny carotid bifurcation lumens from very large backgrounds with no manual intervention. This was the first attempt to use deep learning to achieve carotid bifurcation segmentation in 3D CTA images. Our results indicate that deep learning is a promising method for automatically extracting carotid bifurcation lumens.

Zhou Tianshu, Tan Tao, Pan Xiaoyan, Tang Hui, Li Jingsong

2021-Jan

Automatic lumen segmentation, U-Net, carotid stenosis, dominance of background voxels, large 3D volumes

General General

Assessment of muscle activity using electrical stimulation and mechanomyography: a systematic review.

In Biomedical engineering online

This research has proved that mechanomyographic (MMG) signals can be used for evaluating muscle performance. Stimulation of the lost physiological functions of a muscle using an electrical signal has been determined crucial in clinical and experimental settings in which voluntary contraction fails in stimulating specific muscles. Previous studies have already indicated that characterizing contractile properties of muscles using MMG through neuromuscular electrical stimulation (NMES) showed excellent reliability. Thus, this review highlights the use of MMG signals on evaluating skeletal muscles under electrical stimulation. In total, 336 original articles were identified from the Scopus and SpringerLink electronic databases using search keywords for studies published between 2000 and 2020, and their eligibility for inclusion in this review has been screened using various inclusion criteria. After screening, 62 studies remained for analysis, with two additional articles from the bibliography, were categorized into the following: (1) fatigue, (2) torque, (3) force, (4) stiffness, (5) electrode development, (6) reliability of MMG and NMES approaches, and (7) validation of these techniques in clinical monitoring. This review has found that MMG through NMES provides feature factors for muscle activity assessment, highlighting standardized electromyostimulation and MMG parameters from different experimental protocols. Despite the evidence of mathematical computations in quantifying MMG along with NMES, the requirement of the processing speed, and fluctuation of MMG signals influence the technique to be prone to errors. Interestingly, although this review does not focus on machine learning, there are only few studies that have adopted it as an alternative to statistical analysis in the assessment of muscle fatigue, torque, and force. The results confirm the need for further investigation on the use of sophisticated computations of features of MMG signals from electrically stimulated muscles in muscle function assessment and assistive technology such as prosthetics control.

Uwamahoro Raphael, Sundaraj Kenneth, Subramaniam Indra Devi

2021-Jan-03

Electrical stimulation, Mechanomyography, Muscle activity, Muscle assessment, Muscle mechanics