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

Public Health Public Health

Detecting Tweets Containing Cannabidiol-Related COVID-19 Misinformation Using Transformer Language Models and Warning Letters From Food and Drug Administration: Content Analysis and Identification.

In JMIR infodemiology

BACKGROUND : COVID-19 has introduced yet another opportunity to web-based sellers of loosely regulated substances, such as cannabidiol (CBD), to promote sales under false pretenses of curing the disease. Therefore, it has become necessary to innovate ways to identify such instances of misinformation.

OBJECTIVE : We sought to identify COVID-19 misinformation as it relates to the sales or promotion of CBD and used transformer-based language models to identify tweets semantically similar to quotes taken from known instances of misinformation. In this case, the known misinformation was the publicly available Warning Letters from Food and Drug Administration (FDA).

METHODS : We collected tweets using CBD- and COVID-19-related terms. Using a previously trained model, we extracted the tweets indicating commercialization and sales of CBD and annotated those containing COVID-19 misinformation according to the FDA definitions. We encoded the collection of tweets and misinformation quotes into sentence vectors and then calculated the cosine similarity between each quote and each tweet. This allowed us to establish a threshold to identify tweets that were making false claims regarding CBD and COVID-19 while minimizing the instances of false positives.

RESULTS : We demonstrated that by using quotes taken from Warning Letters issued by FDA to perpetrators of similar misinformation, we can identify semantically similar tweets that also contain misinformation. This was accomplished by identifying a cosine distance threshold between the sentence vectors of the Warning Letters and tweets.

CONCLUSIONS : This research shows that commercial CBD or COVID-19 misinformation can potentially be identified and curbed using transformer-based language models and known prior instances of misinformation. Our approach functions without the need for labeled data, potentially reducing the time at which misinformation can be identified. Our approach shows promise in that it is easily adapted to identify other forms of misinformation related to loosely regulated substances.

Turner Jason, Kantardzic Mehmed, Vickers-Smith Rachel, Brown Andrew G

2023

COVID-19, Twitter, cannabidiol, content analysis, deep learning, health information, infodemic, infodemiology, language model, misinformation, pandemic, sentence vector, social media, transformer

Public Health Public Health

Impact of human mobility on COVID-19 transmission according to mobility distance, locations and demographic factors in the Greater Bay area of China：a population-based study.

In JMIR public health and surveillance

BACKGROUND : Mobility restriction is one of the primary measures used to restrain the spread of COVID-19 in the pandemic all over the world. Governments implemented and relaxed various mobility restriction measures in the absence of evidence for almost three years, which caused severe adverse outcomes in health, society and economy.

OBJECTIVE : This study aims to quantify the impact of mobility reduction on COVID-19 transmission according to mobility distance, locations and demographic factors to identify hotspots of transmission and guide public health policies.

METHODS : Millions of the anonymized, aggregated mobile phone position data between Jan 1 and Feb 24, 2020 was collected for the nine mega cities Greater Bay Area (GBA), China. A generalized linear model (GLM) was established to test the association between mobility volume (number of trips) and COVID-19 transmission. Subgroups analysis was also performed for sex, age, travel locations and travel distance. The statistical interaction terms were included in a variety of models that express different relations between the involved variables.

RESULTS : The GLM analysis demonstrated a significant association between the COVID-19 growth rate ratio (GR) and mobility volume. A stratification analysis revealed a higher effect of mobility volume on the COVID-19 growth rate ratio (GR) among people aged 50-59 years (a decrease of 13.17% for GR per 10% reduction of mobility volume for persons 50-59 years, P <.001) than for other age groups (a decrease of 7.80%, 10.43%, 7.48%, 8.01%, 10.43% for age groups of ≤18, 19-29, 30-39, 40-49, ≥ 60 years, respectively, Pinteraction=.024). The impact of mobility reduction on COVID-19 transmission was higher in transit stations and shopping areas: a decrease of 0.67, 0.53, 0.30, 0.37, 0.44, 0.32 for instantaneous reproduction number R(t) per 10% reduction in mobility volume to transit stations, shopping, work, school, recreation, and other locations, separately (Pinteraction=.016). The association between reduction in mobility volume and COVID-19 transmission was lower with decreasing mobility distance as there was significant interaction between mobility volume and mobility distance on R(t) (Pinteraction<.001). Specifically, the R(t) reduced by 11.97% per 10% reduction of mobility volume when the mobility distance increased to 110% (Spring Festival), by 6.74% when distance remained unchanged and by 1.52% when the distance decreased to 90%.

CONCLUSIONS : The association between mobility reduction and COVID-19 transmission significantly varied by mobility distance, locations and age. The substantially higher impact of mobility volume on COVID-19 transmission in longer travel distance, certain age groups, and for specific travel destinations highlights the potential to optimize the effectiveness of mobility restriction strategies. The results from our study demonstrate the power of having a mobility network using mobile phone data for surveillance that monitor movement at a detailed level to measure the potential impacts of future pandemics.

Xia Jizhe, Yin Kun, Yue Yang, Li Qingquan, Wang Xiling, Hu Dongsheng, Wang Xiong, Du Zhanwei, Cowling Ben J, Chen Erzhen, Zhou Ying

2023-Feb-23

Pathology Pathology

3D bioprinting and the revolution in experimental cancer model systems-A review of developing new models and experiences with in vitro 3D bioprinted breast cancer tissue-mimetic structures.

In Pathology oncology research : POR

Growing evidence propagates those alternative technologies (relevant human cell-based-e.g., organ-on-chips or biofabricated models-or artificial intelligence-combined technologies) that could help in vitro test and predict human response and toxicity in medical research more accurately. In vitro disease model developments have great efforts to create and serve the need of reducing and replacing animal experiments and establishing human cell-based in vitro test systems for research use, innovations, and drug tests. We need human cell-based test systems for disease models and experimental cancer research; therefore, in vitro three-dimensional (3D) models have a renaissance, and the rediscovery and development of these technologies are growing ever faster. This recent paper summarises the early history of cell biology/cellular pathology, cell-, tissue culturing, and cancer research models. In addition, we highlight the results of the increasing use of 3D model systems and the 3D bioprinted/biofabricated model developments. Moreover, we present our newly established 3D bioprinted luminal B type breast cancer model system, and the advantages of in vitro 3D models, especially the bioprinted ones. Based on our results and the reviewed developments of in vitro breast cancer models, the heterogeneity and the real in vivo situation of cancer tissues can be represented better by using 3D bioprinted, biofabricated models. However, standardising the 3D bioprinting methods is necessary for future applications in different high-throughput drug tests and patient-derived tumour models. Applying these standardised new models can lead to the point that cancer drug developments will be more successful, efficient, and consequently cost-effective in the near future.

Sztankovics Dániel, Moldvai Dorottya, Petővári Gábor, Gelencsér Rebeka, Krencz Ildikó, Raffay Regina, Dankó Titanilla, Sebestyén Anna

2023

3D bioprinting, biofabrication, breast cancer, cancer, disease models

Ophthalmology Ophthalmology

Glaucoma and Telemedicine.

In Journal of glaucoma

The coronavirus disease 2019 (COVID-19) pandemic drastically impacted global health, forcing institutions to provide alternative models of safe and reliable health care. In this context, telemedicine has been successfully used to overcome distance barriers and improve access to medical services. Teleglaucoma is the application of telemedicine to screen and monitor glaucoma, a chronic and progressive optic neuropathy. Teleglaucoma screening aims to detect the disease at an earlier stage, especially in high-risk populations and underserved areas, also identifying patients who require more urgent treatment. Teleglaucoma monitoring seeks to provide remote management through virtual clinics, where classical in-person visits are replaced by synchronous data collection (clinical measurements) performed by non-ophthalmologists and asynchronous review (decision-making) by ophthalmologists. This may be employed for low-risk patients with early disease, improving health care logistics, reducing number of face-to-face consultations, and saving time and costs. New technologies may also allow home monitoring of patients in teleglaucoma programs, with addition of artificial intelligence methods, which are expected to increase accuracy of remote glaucoma screening/ monitoring, and to support clinical decision-making. However, for incorporation of teleglaucoma into clinical practice, a complex system for collection, transfer, flow, and interpretation of data is still necessary, in addition to clearer regulatory markers by government agencies and medical entities.

Brandão-de-Resende Camilo, de Alcântara Liliane de Abreu Rosa, Vasconcelos-Santos Daniel Vítor, Diniz-Filho Alberto

2023-Feb-28

General General

Neural text generation in regulatory medical writing.

In Frontiers in pharmacology

Background: A steep increase in new drug applications has increased the overhead of writing technical documents such as medication guides. Natural language processing can contribute to reducing this burden. Objective: To generate medication guides from texts that relate to prescription drug labeling information. Materials and Methods: We collected official drug label information from the DailyMed website. We focused on drug labels containing medication guide sections to train and test our model. To construct our training dataset, we aligned "source" text from the document with similar "target" text from the medication guide using three families of alignment techniques: global, manual, and heuristic alignment. The resulting source-target pairs were provided as input to a Pointer Generator Network, an abstractive text summarization model. Results: Global alignment produced the lowest ROUGE scores and relatively poor qualitative results, as running the model frequently resulted in mode collapse. Manual alignment also resulted in mode collapse, albeit higher ROUGE scores than global alignment. Within the family of heuristic alignment approaches, we compared different methods and found BM25-based alignments to produce significantly better summaries (at least 6.8 ROUGE points above the other techniques). This alignment surpassed both the global and manual alignments in terms of ROUGE and qualitative scoring. Conclusion: The results of this study indicate that a heuristic approach to generating inputs for an abstractive summarization model increased ROUGE scores, compared to a global or manual approach when automatically generating biomedical text. Such methods hold the potential to significantly reduce the manual labor burden in medical writing and related disciplines.

Meyer Claudia, Adkins Daniel, Pal Koyena, Galici Ruggero, Garcia-Agundez Augusto, Eickhoff Carsten

2023

abstractive summarization, artificial intelligence-AI, drug labeling, medication guide, natural language generation, pointer generator network

Public Health Public Health

Artificial Intelligence Functionalities During the COVID-19 Pandemic.

In Disaster medicine and public health preparedness

BACKGROUND : The coronavirus disease 2019 (COVID-19) pandemic has led us to use virtual solutions and emerging technologies such as artificial intelligence (AI). Recent studies have clearly demonstrated the role of AI in health care and medical practice; however, a comprehensive review can identify potential yet not fulfilled functionalities of such technologies in pandemics. Therefore, this scoping review study aims at assessing AI functionalities in the COVID-19 pandemic in 2022.

METHODS : A systematic search was carried out in PubMed, Cochran Library, Scopus, Science Direct, ProQuest, and Web of Science from 2019 to May 9, 2022. Researchers selected the articles according to the search keywords. Finally, the articles mentioning the functionalities of AI in the COVID-19 pandemic were evaluated. Two investigators performed this process.

RESULTS : Initial search resulted in 9123 articles. After reviewing the title, abstract, and full text of these articles, and applying the inclusion and exclusion criteria, 4 articles were selectd for the final analysis. All 4 were cross-sectional studies. Two studies (50%) were performed in the United States, 1 (25%) in Israel, and 1 (25%) in Saudi Arabia. They covered the functionalities of AI in the prediction, detection, and diagnosis of COVID-19.

CONCLUSIONS : To the extent of the researchers' knowledge, this study is the first scoping review that assesses the AI functionalities in the COVID-19 pandemic. Health-care organizations need decision support technologies and evidence-based apparatuses that can perceive, think, and reason not dissimilar to human beings. Potential functionalities of such technologies can be used to predict mortality, detect, screen, and trace current and former patients, analyze health data, prioritize high-risk patients, and better allocate hospital resources in pandemics, and generally in health-care settings.

Ahmadi Marzaleh Milad, Peyravi Mahmoudreza, Mousavi Shahrokh, Sarpourian Fatemeh, Seyedi Milad, Shalyari Naseh

2023-Feb-27

COVID-19, artificial intelligence, deep learning, machine learning, neural networks