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

General General

Efficient Detection of Knee Anterior Cruciate Ligament from Magnetic Resonance Imaging Using Deep Learning Approach.

In Diagnostics (Basel, Switzerland)

The most commonly injured ligament in the human body is an anterior cruciate ligament (ACL). ACL injury is standard among the football, basketball and soccer players. The study aims to detect anterior cruciate ligament injury in an early stage via efficient and thorough automatic magnetic resonance imaging without involving radiologists, through a deep learning method. The proposed approach in this paper used a customized 14 layers ResNet-14 architecture of convolutional neural network (CNN) with six different directions by using class balancing and data augmentation. The performance was evaluated using accuracy, sensitivity, specificity, precision and F1 score of our customized ResNet-14 deep learning architecture with hybrid class balancing and real-time data augmentation after 5-fold cross-validation, with results of 0.920%, 0.916%, 0.946%, 0.916% and 0.923%, respectively. For our proposed ResNet-14 CNN the average area under curves (AUCs) for healthy tear, partial tear and fully ruptured tear had results of 0.980%, 0.970%, and 0.999%, respectively. The proposing diagnostic results indicated that our model could be used to detect automatically and evaluate ACL injuries in athletes using the proposed deep-learning approach.

Awan Mazhar Javed, Rahim Mohd Shafry Mohd, Salim Naomie, Mohammed Mazin Abed, Garcia-Zapirain Begonya, Abdulkareem Karrar Hameed


MRI, anterior cruciate ligament, artificial intelligence, augmentation, classification, convolutional neural network, detection, healthcare, knee injury, residual network

General General

Fighting viruses with materials science: Prospects for antivirus surfaces, drug delivery systems and artificial intelligence.

In Dental materials : official publication of the Academy of Dental Materials

OBJECTIVE : Viruses on environmental surfaces, in saliva and other body fluids represent risk of contamination for general population and healthcare professionals. The development of vaccines and medicines is costly and time consuming. Thus, the development of novel materials and technologies to decrease viral availability, viability, infectivity, and to improve therapeutic outcomes can positively impact the prevention and treatment of viral diseases.

METHODS : Herein, we discuss (a) interaction mechanisms between viruses and materials, (b) novel strategies to develop materials with antiviral properties and oral antiviral delivery systems, and (c) the potential of artificial intelligence to design and optimize preventive measures and therapeutic regimen.

RESULTS : The mechanisms of viral adsorption on surfaces are well characterized but no major breakthrough has become clinically available. Materials with fine-tuned physical and chemical properties have the potential to compromise viral availability and stability. Emerging strategies using oral antiviral delivery systems and artificial intelligence can decrease infectivity and improve antiviral therapies.

SIGNIFICANCE : Emerging viral infections are concerning due to risk of mortality, as well as psychological and economic impacts. Materials science emerges for the development of novel materials and technologies to diminish viral availability, infectivity, and to enable enhanced preventive and therapeutic strategies, for the safety and well-being of humankind.

Rosa Vinicius, Ho Dean, Sabino-Silva Robinson, Siqueira Walter L, Silikas Nikolaos


COVID-19, Coating, Coronavirus, Diagnostic, Infection, Nanomaterial, Nanotechnology, Pandemic, Saliva, Vaccine

Radiology Radiology

Cone-beam CT image quality improvement using Cycle-Deblur consistent adversarial networks (Cycle-Deblur GAN) for chest CT imaging in breast cancer patients.

In Scientific reports ; h5-index 158.0

Cone-beam computed tomography (CBCT) integrated with a linear accelerator is widely used to increase the accuracy of radiotherapy and plays an important role in image-guided radiotherapy (IGRT). For comparison with fan-beam computed tomography (FBCT), the image quality of CBCT is indistinct due to X-ray scattering, noise, and artefacts. We proposed a deep learning model, "Cycle-Deblur GAN", combined with CycleGAN and Deblur-GAN models to improve the image quality of chest CBCT images. The 8706 CBCT and FBCT image pairs were used for training, and 1150 image pairs were used for testing in deep learning. The generated CBCT images from the Cycle-Deblur GAN model demonstrated closer CT values to FBCT in the lung, breast, mediastinum, and sternum compared to the CycleGAN and RED-CNN models. The quantitative evaluations of MAE, PSNR, and SSIM for CBCT generated from the Cycle-Deblur GAN model demonstrated better results than the CycleGAN and RED-CNN models. The Cycle-Deblur GAN model improved image quality and CT-value accuracy and preserved structural details for chest CBCT images.

Tien Hui-Ju, Yang Hsin-Chih, Shueng Pei-Wei, Chen Jyh-Cheng


Public Health Public Health

Comparison of deep learning with traditional models to predict preventable acute care use and spending among heart failure patients.

In Scientific reports ; h5-index 158.0

Recent health reforms have created incentives for cardiologists and accountable care organizations to participate in value-based care models for heart failure (HF). Accurate risk stratification of HF patients is critical to efficiently deploy interventions aimed at reducing preventable utilization. The goal of this paper was to compare deep learning approaches with traditional logistic regression (LR) to predict preventable utilization among HF patients. We conducted a prognostic study using data on 93,260 HF patients continuously enrolled for 2-years in a large U.S. commercial insurer to develop and validate prediction models for three outcomes of interest: preventable hospitalizations, preventable emergency department (ED) visits, and preventable costs. Patients were split into training, validation, and testing samples. Outcomes were modeled using traditional and enhanced LR and compared to gradient boosting model and deep learning models using sequential and non-sequential inputs. Evaluation metrics included precision (positive predictive value) at k, cost capture, and Area Under the Receiver operating characteristic (AUROC). Deep learning models consistently outperformed LR for all three outcomes with respect to the chosen evaluation metrics. Precision at 1% for preventable hospitalizations was 43% for deep learning compared to 30% for enhanced LR. Precision at 1% for preventable ED visits was 39% for deep learning compared to 33% for enhanced LR. For preventable cost, cost capture at 1% was 30% for sequential deep learning, compared to 18% for enhanced LR. The highest AUROCs for deep learning were 0.778, 0.681 and 0.727, respectively. These results offer a promising approach to identify patients for targeted interventions.

Lewis Maor, Elad Guy, Beladev Moran, Maor Gal, Radinsky Kira, Hermann Dor, Litani Yoav, Geller Tal, Pines Jesse M, Shapiro Nathan L, Figueroa Jose F


General General

Embeddings from deep learning transfer GO annotations beyond homology.

In Scientific reports ; h5-index 158.0

Knowing protein function is crucial to advance molecular and medical biology, yet experimental function annotations through the Gene Ontology (GO) exist for fewer than 0.5% of all known proteins. Computational methods bridge this sequence-annotation gap typically through homology-based annotation transfer by identifying sequence-similar proteins with known function or through prediction methods using evolutionary information. Here, we propose predicting GO terms through annotation transfer based on proximity of proteins in the SeqVec embedding rather than in sequence space. These embeddings originate from deep learned language models (LMs) for protein sequences (SeqVec) transferring the knowledge gained from predicting the next amino acid in 33 million protein sequences. Replicating the conditions of CAFA3, our method reaches an Fmax of 37 ± 2%, 50 ± 3%, and 57 ± 2% for BPO, MFO, and CCO, respectively. Numerically, this appears close to the top ten CAFA3 methods. When restricting the annotation transfer to proteins with < 20% pairwise sequence identity to the query, performance drops (Fmax BPO 33 ± 2%, MFO 43 ± 3%, CCO 53 ± 2%); this still outperforms naïve sequence-based transfer. Preliminary results from CAFA4 appear to confirm these findings. Overall, this new concept is likely to change the annotation of proteins, in particular for proteins from smaller families or proteins with intrinsically disordered regions.

Littmann Maria, Heinzinger Michael, Dallago Christian, Olenyi Tobias, Rost Burkhard


General General

Modeling of trees failure under windstorm in harvested Hyrcanian forests using machine learning techniques.

In Scientific reports ; h5-index 158.0

In managed forests, windstorm disturbances reduce the yield of timber by imposing the costs of unscheduled clear-cutting or thinning operations. Hyrcanian forests are affected by permanent winds, with more than 100 km/h which cause damage forest trees and in result of the tree harvesting and gap creation in forest stands, many trees failure accidents happen annually. Using machine learning approaches, we aimed to compare the multi-layer perceptron (MLP) neural network, radial basis function neural network (RBFNN) and support vector machine (SVM) models for identifying susceptible trees in windstorm disturbances. Therefore, we recorded 15 variables in 600 sample plots which are divided into two categories: 1. Stand variables and 2.Tree variables. We developed the tree failure model (TFM) by artificial intelligence techniques such as MLP, RBFNN, and SVM. The MLP model represents the highest accuracy of target trees classification in training (100%), test (93.3%) and all data sets (97.7%). The values of the mean of trees height, tree crown diameter, target tree height are prioritized respectively as the most significant inputs which influence tree susceptibility in windstorm disturbances. The results of MLP modeling defined TFMmlp as a comparative impact assessment model in susceptible tree identification in Hyrcanian forests where the tree failure is in result of the susceptibility of remained trees after wood harvesting. The TFMmlp is applicable in Hyrcanian forest management planning for wood harvesting to decrease the rate of tree failure after wood harvesting and a tree cutting plan could be modified based on designed environmental decision support system tool to reduce the risk of trees failure in wind circulations.

Jahani Ali, Saffariha Maryam