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

Public Health Public Health

Depression, cardiometabolic disease, and their co-occurrence after childhood maltreatment: an individual participant data meta-analysis including over 200,000 participants.

In BMC medicine ; h5-index 89.0

BACKGROUND : Childhood maltreatment is associated with depression and cardiometabolic disease in adulthood. However, the relationships with these two diseases have so far only been evaluated in different samples and with different methodology. Thus, it remains unknown how the effect sizes magnitudes for depression and cardiometabolic disease compare with each other and whether childhood maltreatment is especially associated with the co-occurrence ("comorbidity") of depression and cardiometabolic disease. This pooled analysis examined the association of childhood maltreatment with depression, cardiometabolic disease, and their comorbidity in adulthood.

METHODS : We carried out an individual participant data meta-analysis on 13 international observational studies (N = 217,929). Childhood maltreatment comprised self-reports of physical, emotional, and/or sexual abuse before 18 years. Presence of depression was established with clinical interviews or validated symptom scales and presence of cardiometabolic disease with self-reported diagnoses. In included studies, binomial and multinomial logistic regressions estimated sociodemographic-adjusted associations of childhood maltreatment with depression, cardiometabolic disease, and their comorbidity. We then additionally adjusted these associations for lifestyle factors (smoking status, alcohol consumption, and physical activity). Finally, random-effects models were used to pool these estimates across studies and examined differences in associations across sex and maltreatment types.

RESULTS : Childhood maltreatment was associated with progressively higher odds of cardiometabolic disease without depression (OR [95% CI] = 1.27 [1.18; 1.37]), depression without cardiometabolic disease (OR [95% CI] = 2.68 [2.39; 3.00]), and comorbidity between both conditions (OR [95% CI] = 3.04 [2.51; 3.68]) in adulthood. Post hoc analyses showed that the association with comorbidity was stronger than with either disease alone, and the association with depression was stronger than with cardiometabolic disease. Associations remained significant after additionally adjusting for lifestyle factors, and were present in both males and females, and for all maltreatment types.

CONCLUSIONS : This meta-analysis revealed that adults with a history of childhood maltreatment suffer more often from depression and cardiometabolic disease than their non-exposed peers. These adults are also three times more likely to have comorbid depression and cardiometabolic disease. Childhood maltreatment may therefore be a clinically relevant indicator connecting poor mental and somatic health. Future research should investigate the potential benefits of early intervention in individuals with a history of maltreatment on their distal mental and somatic health (PROSPERO CRD42021239288).

Souama Camille, Lamers Femke, Milaneschi Yuri, Vinkers Christiaan H, Defina Serena, Garvert Linda, Stein Frederike, Woofenden Tom, Brosch Katharina, Dannlowski Udo, Galenkamp Henrike, de Graaf Ron, Jaddoe Vincent W V, Lok Anja, van Rijn Bas B, Völzke Henry, Cecil Charlotte A M, Felix Janine F, Grabe Hans J, Kircher Tilo, Lekadir Karim, Have Margreet Ten, Walton Esther, Penninx Brenda W J H

2023-Mar-13

Adverse childhood experiences, Cardiovascular diseases, Child abuse, Childhood maltreatment, Comorbidity, Depression, Depressive disorder, Diabetes mellitus, Meta-analysis

Ophthalmology Ophthalmology

Implementation of deep learning artificial intelligence in vision-threatening disease screenings for an underserved community during COVID-19.

In Journal of telemedicine and telecare ; h5-index 28.0

INTRODUCTION : Age-related macular degeneration, diabetic retinopathy, and glaucoma are vision-threatening diseases that are leading causes of vision loss. Many studies have validated deep learning artificial intelligence for image-based diagnosis of vision-threatening diseases. Our study prospectively investigated deep learning artificial intelligence applications in student-run non-mydriatic screenings for an underserved, primarily Hispanic community during COVID-19.

METHODS : Five supervised student-run community screenings were held in West New York, New Jersey. Participants underwent non-mydriatic 45-degree retinal imaging by medical students. Images were uploaded to a cloud-based deep learning artificial intelligence for vision-threatening disease referral. An on-site tele-ophthalmology grader and remote clinical ophthalmologist graded images, with adjudication by a senior ophthalmologist to establish the gold standard diagnosis, which was used to assess the performance of deep learning artificial intelligence.

RESULTS : A total of 385 eyes from 195 screening participants were included (mean age 52.43  ±  14.5 years, 40.0% female). A total of 48 participants were referred for at least one vision-threatening disease. Deep learning artificial intelligence marked 150/385 (38.9%) eyes as ungradable, compared to 10/385 (2.6%) ungradable as per the human gold standard (p < 0.001). Deep learning artificial intelligence had 63.2% sensitivity, 94.5% specificity, 32.0% positive predictive value, and 98.4% negative predictive value in vision-threatening disease referrals. Deep learning artificial intelligence successfully referred all 4 eyes with multiple vision-threatening diseases. Deep learning artificial intelligence graded images (35.6  ±  13.3 s) faster than the tele-ophthalmology grader (129  ±  41.0) and clinical ophthalmologist (68  ±  21.9, p < 0.001).

DISCUSSION : Deep learning artificial intelligence can increase the efficiency and accessibility of vision-threatening disease screenings, particularly in underserved communities. Deep learning artificial intelligence should be adaptable to different environments. Consideration should be given to how deep learning artificial intelligence can best be utilized in a real-world application, whether in computer-aided or autonomous diagnosis.

Zhu Aretha, Tailor Priya, Verma Rashika, Zhang Isis, Schott Brian, Ye Catherine, Szirth Bernard, Habiel Miriam, Khouri Albert S

2023-Mar-13

COVID-19, Tele-ophthalmology, deep learning artificial intelligence, vision screenings

Radiology Radiology

Medical Phrase Grounding with Region-Phrase Context Contrastive Alignment

ArXiv Preprint

Medical phrase grounding (MPG) aims to locate the most relevant region in a medical image, given a phrase query describing certain medical findings, which is an important task for medical image analysis and radiological diagnosis. However, existing visual grounding methods rely on general visual features for identifying objects in natural images and are not capable of capturing the subtle and specialized features of medical findings, leading to sub-optimal performance in MPG. In this paper, we propose MedRPG, an end-to-end approach for MPG. MedRPG is built on a lightweight vision-language transformer encoder and directly predicts the box coordinates of mentioned medical findings, which can be trained with limited medical data, making it a valuable tool in medical image analysis. To enable MedRPG to locate nuanced medical findings with better region-phrase correspondences, we further propose Tri-attention Context contrastive alignment (TaCo). TaCo seeks context alignment to pull both the features and attention outputs of relevant region-phrase pairs close together while pushing those of irrelevant regions far away. This ensures that the final box prediction depends more on its finding-specific regions and phrases. Experimental results on three MPG datasets demonstrate that our MedRPG outperforms state-of-the-art visual grounding approaches by a large margin. Additionally, the proposed TaCo strategy is effective in enhancing finding localization ability and reducing spurious region-phrase correlations.

Zhihao Chen, Yang Zhou, Anh Tran, Junting Zhao, Liang Wan, Gideon Ooi, Lionel Cheng, Choon Hua Thng, Xinxing Xu, Yong Liu, Huazhu Fu

2023-03-14

General General

Heating Up: How Early Twitter Marketing Gave Rise to Organic Word-of-Mouth About Heated Tobacco Products.

In Social media + society

Social media are an important marketing platform for emerging tobacco products. Heated tobacco products (HTPs) have been introduced in a limited number of local test markets in the United States as potentially reduced-exposure tobacco products. HTPs can be used to heat tobacco as well as marijuana. However, due to growing digital media promotion of these products, it is possible that public knowledge and purchasing opportunities extend beyond test markets. Research on HTP social media promotion is sparse. The objective of the present study is to assess the amount and characterize the content of HTP-related messages on Twitter. We used keyword rules to collect HTP-related posts from the Twitter Historical Powertrack from 1 August 2016 to 31 July 2021. Posts were coded for type (organic, commercial), promotional strategies (e.g., discounts, event promotion), and marijuana co-use mentions using a combination of machine learning methods and human coding. Keyword filters captured 121,012 relevant tweets posted over the period of data collection, with 46,013 (38.02%) tweets featuring commercial content. Findings revealed that there was a two-fold increase in the monthly volume of messages from August 2016 to July 2021. The proportion of organic tweets increased from 29% of all tweets in August 2016 to 73.5% in July 2021. Approximately 20.6% of tweets included mentions of marijuana, and 5,243 posts (4.3%) contained links to online retailers. Promotional tweets featured event promotion, discounts, reduced harm appeals, and fashion appeals. Tobacco control and substance use prevention initiatives should include efforts to monitor the role of social media in promoting organic word-of-mouth and normalizing novel tobacco products.

Kostygina Ganna, Tran Hy, Kim Yoonsang, Czaplicki Lauren, Kierstead Elexis, Kreslake Jennifer, Emery Sherry, Schillo Barbara

2022

advertising and promotion, social media

General General

Cognitive rehabilitation in multiple sclerosis: Three digital ingredients to address current and future priorities.

In Frontiers in human neuroscience ; h5-index 79.0

Multiple sclerosis (MS) is a neurological chronic disease with autoimmune demyelinating lesions and one of the most common disability causes in young adults. People with MS (PwMS) experience cognitive impairments (CIs) and clinical evidence shows their presence during all MS stages even in the absence of other symptoms. Cognitive rehabilitation (CR) aims at reducing CI and improving PwMS' awareness of cognitive difficulties faced in their daily living. More defined cognitive profiles, easier treatment access and the need to transfer intervention effects into everyday life activities are aims of utmost relevance for CR in MS. Currently, advanced technologies may pave the way to rethink CR in MS to address the priority of more personalized and effective, accessible and ecological interventions. For this purpose, digital twins, tele-cognitive-rehabilitation and metaverse are the main candidate digital ingredients. Based on scientific evidences, we propose digital twin technology to enhance MS cognitive phenotyping; tele-cognitive-rehabilitation to make feasible the cognitive intervention access to a larger number of PwMS; and metaverse to represent the best choice to train real-world dual- and multi-tasking deficits in virtual daily life environments. Moreover, multi-domain high-frequency big-data collected through tele-cognitive-assessment, tele-cognitive-rehabilitation, and metaverse may be merged to refine artificial intelligence algorithms and obtain increasingly detailed patient's cognitive profile in order to enhance intervention personalization. Here, we present how these digital ingredients and their integration could be crucial to address the current and future needs of CR facilitating the early detection of subtle CI and the delivery of increasingly effective treatments.

Tacchino Andrea, Podda Jessica, Bergamaschi Valeria, Pedullà Ludovico, Brichetto Giampaolo

2023

cognitive deficit, cognitive disorder, cognitive impairment, digital twin, dual-task, metaverse, multiple sclerosis, telerehabilitation

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