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

Public Health Public Health

IPs-GRUAtt: An attention-based bidirectional gated recurrent unit network for predicting phosphorylation sites of SARS-CoV-2 infection.

In Molecular therapy. Nucleic acids

The global pandemic of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has generated tremendous concern and poses a serious threat to international public health. Phosphorylation is a common post-translational modification affecting many essential cellular processes and is inextricably linked to SARS-CoV-2 infection. Hence, accurate identification of phosphorylation sites will be helpful to understand the mechanisms of SARS-CoV-2 infection and mitigate the ongoing COVID-19 pandemic. In the present study, an attention-based bidirectional gated recurrent unit network, called IPs-GRUAtt, was proposed to identify phosphorylation sites in SARS-CoV-2-infected host cells. Comparative results demonstrated that IPs-GRUAtt surpassed both state-of-the-art machine-learning methods and existing models for identifying phosphorylation sites. Moreover, the attention mechanism made IPs-GRUAtt able to extract the key features from protein sequences. These results demonstrated that the IPs-GRUAtt is a powerful tool for identifying phosphorylation sites. For facilitating its academic use, a freely available online web server for IPs-GRUAtt is provided at http://cbcb.cdutcm.edu.cn/phosphory/.

Zhang Guiyang, Tang Qiang, Feng Pengmian, Chen Wei

2023-Jun-13

MT: Bioinformatics, SARS-CoV-2, attention mechanism, bidirectional gated recurrent unit, deep learning, interpretation, phosphorylation

General General

Tensor-based Multimodal Learning for Prediction of Pulmonary Arterial Wedge Pressure from Cardiac MRI

ArXiv Preprint

Heart failure is a serious and life-threatening condition that can lead to elevated pressure in the left ventricle. Pulmonary Arterial Wedge Pressure (PAWP) is an important surrogate marker indicating high pressure in the left ventricle. PAWP is determined by Right Heart Catheterization (RHC) but it is an invasive procedure. A non-invasive method is useful in quickly identifying high-risk patients from a large population. In this work, we develop a tensor learning-based pipeline for identifying PAWP from multimodal cardiac Magnetic Resonance Imaging (MRI). This pipeline extracts spatial and temporal features from high-dimensional scans. For quality control, we incorporate an epistemic uncertainty-based binning strategy to identify poor-quality training samples. To improve the performance, we learn complementary information by integrating features from multimodal data: cardiac MRI with short-axis and four-chamber views, and Electronic Health Records. The experimental analysis on a large cohort of $1346$ subjects who underwent the RHC procedure for PAWP estimation indicates that the proposed pipeline has a diagnostic value and can produce promising performance with significant improvement over the baseline in clinical practice (i.e., $\Delta$AUC $=0.10$, $\Delta$Accuracy $=0.06$, and $\Delta$MCC $=0.39$). The decision curve analysis further confirms the clinical utility of our method.

Prasun C. Tripathi, Mohammod N. I. Suvon, Lawrence Schobs, Shuo Zhou, Samer Alabed, Andrew J. Swift, Haiping Lu

2023-03-14

oncology Oncology

Preoperative Prediction and Identification of Extracapsular Extension in Head and Neck Cancer Patients: Progress and Potential.

In Cureus

Background This study aimed to demonstrate both the potential and development progress in the identification of extracapsular nodal extension in head and neck cancer patients prior to surgery. Methodology A deep learning model has been developed utilizing multilayer gradient mapping-guided explainable network architecture involving a volume extractor. In addition, the gradient-weighted class activation mapping approach has been appropriated to generate a heatmap of anatomic regions indicating why the algorithm predicted extension or not. Results The prediction model shows excellent performance on the testing dataset with high values of accuracy, the area under the curve, sensitivity, and specificity of 0.926, 0.945, 0.924, and 0.930, respectively. The heatmap results show potential usefulness for some select patients but indicate the need for further training as the results may be misleading for other patients. Conclusions This work demonstrates continued progress in the identification of extracapsular nodal extension in diagnostic computed tomography prior to surgery. Continued progress stands to see the obvious potential realized where not only can unnecessary multimodality therapy be avoided but necessary therapy can be guided on a patient-specific level with information that currently is not available until postoperative pathology is complete.

Duggar William N, Vengaloor Thomas Toms, Wang Yibin, Rahman Abdur, Wang Haifeng, Roberts Paul R, Bian Linkan, Gatewood Ronald T, Vijayakumar Srinivasan

2023-Feb

artificial intelligence, deep learning, extracapsular extension, head and neck squamous cell carcinoma, model explainability, preoperative

General General

Distributed Caregiving for Cognitively Impaired Individuals: A Case Report.

In Cureus

Many caregivers of people with cognitive impairment spend a significant amount of their time helping patients with instrumental daily functions. Distributed caregiving is an innovative model designed to reduce an individual caregiver's time burden and increase the likelihood of continued independent living for the patient. Echo Show and Google Home platforms were used to enable the participation of remote family members in caregiving, specifically the socialization and entertainment of a person with cognitive impairment. Caregiver interviews, review of medical records, and case study analysis were used to measure caregiver burden, after distributing some components of caregiving to distant family members with human-in-the-loop artificial intelligence. This case explores the use of Alexa, Echo Show, and other commercial technologies in the management of a patient with cognitive impairment. The human-in-the-loop system introduced in this case study is a creative, accessible, low-cost, and sustainable way to potentially reduce caregiver burden and improve patient outcomes with targeted intervention. Targeted distributed caregiving reduced time spent in caregiving, reduced caregiver guilt and frustration, improved patient's compliance with requests for behavior changes (e.g., voiding before leaving the house), and improved the relationship between the caregiver and the person with cognitive impairment. This case study demonstrates how distributed caregiving, including human-in-the-loop artificial intelligence, can lead to better use of technology in reducing the social isolation of persons with cognitive impairment and in reducing caregiver burden.

Alemi Yara, Loughman Blaise, Uriyo Maria

2023-Feb

care navigation, caregiver burden, case study, cognitive impairment, home platforms

General General

Sprinting performance of individuals with unilateral transfemoral amputation: compensation strategies for lower limb coordination.

In Royal Society open science

Understanding the sprinting patterns of individuals with unilateral transfemoral amputation (uTFA) is important for designing improved running-specific prostheses and for prosthetic gait rehabilitation. Continuous relative phase (CRP) analysis acquires clues from movement kinematics and obtains insights into the sprinting coordination of individuals with uTFA. Seven individuals with uTFA sprinted on a 40 m runway. The spatio-temporal parameters, joint and segment angles of the lower limbs were obtained, and CRP analysis was performed on thigh-shank and shank-foot couplings. Subsequently, the asymmetry ratios of the parameters were calculated. Statistical analyses were performed between the lower limbs. Significant differences in the stance time, stance phase percentage, ankle joint angles and CRP of the shank-foot coupling (p < 0.05) were observed between the lower limbs. The primary contributor to these differences could be the structural differences between the lower limbs. Despite the presence of different coordination features in the stance and swing phases between the lower limbs, no significant difference in the coordination patterns of the thigh-shank coupling was observed. This may be a compensation strategy for achieving coordination patterns with improved symmetry between the lower limbs. The results of this study provide novel insights into the sprinting movement patterns of individuals with uTFA.

Hu Mingyu, Kobayashi Toshiki, Hisano Genki, Murata Hiroto, Ichimura Daisuke, Hobara Hiroaki

2023-Mar

continuous relative phase, gait, prosthetic, run, running-specific prosthesis

General General

Did AI get more negative recently?

In Royal Society open science

In this paper, we classify scientific articles in the domain of natural language processing (NLP) and machine learning (ML), as core subfields of artificial intelligence (AI), into whether (i) they extend the current state-of-the-art by the introduction of novel techniques which beat existing models or whether (ii) they mainly criticize the existing state-of-the-art, i.e. that it is deficient with respect to some property (e.g. wrong evaluation, wrong datasets, misleading task specification). We refer to contributions under (i) as having a 'positive stance' and contributions under (ii) as having a 'negative stance' (to related work). We annotate over 1.5 k papers from NLP and ML to train a SciBERT-based model to automatically predict the stance of a paper based on its title and abstract. We then analyse large-scale trends on over 41 k papers from the last approximately 35 years in NLP and ML, finding that papers have become substantially more positive over time, but negative papers also got more negative and we observe considerably more negative papers in recent years. Negative papers are also more influential in terms of citations they receive.

Beese Dominik, Altunbaş Begüm, Güzeler Görkem, Eger Steffen

2023-Mar

SciBERT, citation sentiment, natural language processing, negativity, science-of-science, trend prediction