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

Ophthalmology Ophthalmology

Role of artificial intelligence in determining factors impacting patients' refractive surgery decisions.

In Indian journal of ophthalmology

PURPOSE : To create a predictive model using artificial intelligence (AI) and assess if available data from patients' registration records can help in predicting definitive endpoints such as the probability of patients signing up for refractive surgery.

METHODS : This was a retrospective analysis. Electronic health records data of 423 patients presenting to the refractive surgery department were incorporated into models using multivariable logistic regression, decision trees classifier, and random forest (RF). Mean area under the receiver operating characteristic curve (ROC-AUC), sensitivity (Se), specificity (Sp), classification accuracy, precision, recall, and F1-score were calculated for each model to evaluate performance.

RESULTS : The RF classifier provided the best output among the various models, and the top variables identified in this study by the RF classifier excluding income were insurance, time spent in the clinic, age, occupation, residence, source of referral, and so on. About 93% of the cases that did undergo refractive surgery were correctly predicted as having undergone refractive surgery. The AI model achieved an ROC-AUC of 0.945 with an Se of 88% and Sp of 92.5%.

CONCLUSION : This study demonstrated the importance of stratification and identifying various factors using an AI model which could impact patients' decisions while selecting a refractive surgery. Eye centers can build specialized prediction profiles across disease categories and may allow for the identification of prospective obstacles in the patient's decision-making process, as well as strategies for dealing with them.

Kundu Gairik, Virani Imranali, Shetty Rohit, Khamar Pooja, Nuijts Rudy M M A

2023-Mar

Artificial intelligence, machine learning, ophthalmologic surgical procedures, predictive analysis

General General

Rough-set based learning: Assessing patterns and predictability of anxiety, depression, and sleep scores associated with the use of cannabinoid-based medicine during COVID-19.

In Frontiers in artificial intelligence

Recently, research is emerging highlighting the potential of cannabinoids' beneficial effects related to anxiety, mood, and sleep disorders as well as pointing to an increased use of cannabinoid-based medicines since COVID-19 was declared a pandemic. The objective of this research is 3 fold: i) to evaluate the relationship of the clinical delivery of cannabinoid-based medicine for anxiety, depression and sleep scores by utilizing machine learning specifically rough set methods; ii) to discover patterns based on patient features such as specific cannabinoid recommendations, diagnosis information, decreasing/increasing levels of clinical assessment tools (CAT) scores over a period of time; and iii) to predict whether new patients could potentially experience either an increase or decrease in CAT scores. The dataset for this study was derived from patient visits to Ekosi Health Centres, Canada over a 2 year period including the COVID timeline. Extensive pre-processing and feature engineering was performed. A class feature indicative of their progress or lack thereof due to the treatment received was introduced. Six Rough/Fuzzy-Rough classifiers as well as Random Forest and RIPPER classifiers were trained on the patient dataset using a 10-fold stratified CV method. The highest overall accuracy, sensitivity and specificity measures of over 99% was obtained using the rule-based rough-set learning model. In this study, we have identified rough-set based machine learning model with high accuracy that could be utilized for future studies regarding cannabinoids and precision medicine.

Ramanna Sheela, Ashrafi Negin, Loster Evan, Debroni Karen, Turner Shelley

2023

cannabinoid medicine, electronic health records, machine learning, mental health, rough sets, rough-fuzzy sets

General General

Recent deep learning models for dementia as point-of-care testing: Potential for early detection.

In Intractable & rare diseases research

Deep learning has been intensively researched over the last decade, yielding several new models for natural language processing, images, speech and time series processing that have dramatically improved performance. This wave of technological developments in deep learning is also spreading to medicine. The effective use of deep learning in medicine is concentrated in diagnostic imaging-related applications, but deep learning has the potential to lead to early detection and prevention of diseases. Physical aspects of disease that went unnoticed can now be used in diagnosis with deep learning. In particular, deep learning models for the early detection of dementia have been proposed to predict cognitive function based on various information such as blood test results, speech, and the appearance of the face, where the effects of dementia can be seen. Deep learning is a useful diagnostic tool, as it has the potential to detect diseases early based on trivial aspects before clear signs of disease appear. The ability to easily make a simple diagnosis based on information such as blood test results, voice, pictures of the body, and lifestyle is a method suited to point-of-cate testing, which requires immediate testing at the desired time and place. Over the past few years, the process of predicting disease can now be visualized using deep learning, providing insights into new methods of diagnosis.

Karako Kenji, Song Peipei, Chen Yu

2023-Feb

deep learning, dementia, point-of-cate testing, prediction

General General

Polymer Informatics at Scale with Multitask Graph Neural Networks.

In Chemistry of materials : a publication of the American Chemical Society

Artificial intelligence-based methods are becoming increasingly effective at screening libraries of polymers down to a selection that is manageable for experimental inquiry. The vast majority of presently adopted approaches for polymer screening rely on handcrafted chemostructural features extracted from polymer repeat units-a burdensome task as polymer libraries, which approximate the polymer chemical search space, progressively grow over time. Here, we demonstrate that directly "machine learning" important features from a polymer repeat unit is a cheap and viable alternative to extracting expensive features by hand. Our approach-based on graph neural networks, multitask learning, and other advanced deep learning techniques-speeds up feature extraction by 1-2 orders of magnitude relative to presently adopted handcrafted methods without compromising model accuracy for a variety of polymer property prediction tasks. We anticipate that our approach, which unlocks the screening of truly massive polymer libraries at scale, will enable more sophisticated and large scale screening technologies in the field of polymer informatics.

Gurnani Rishi, Kuenneth Christopher, Toland Aubrey, Ramprasad Rampi

2023-Feb-28

Cardiology Cardiology

A multi-use deep learning method for CITE-seq and single-cell RNA-seq data integration with cell surface protein prediction and imputation.

In Nature machine intelligence

CITE-seq, a single-cell multi-omics technology that measures RNA and protein expression simultaneously in single cells, has been widely applied in biomedical research, especially in immune related disorders and other diseases such as influenza and COVID-19. Despite the proliferation of CITE-seq, it is still costly to generate such data. Although data integration can increase information content, this raises computational challenges. First, combining multiple datasets is prone to batch effects that need to be addressed. Secondly, it is difficult to combine multiple CITE-seq datasets because the protein panels in different datasets may only partially overlap. Integrating multiple CITE-seq and single-cell RNA-seq (scRNA-seq) datasets is important because this allows the utilization of as many data as possible to uncover cell population heterogeneity. To overcome these challenges, we present sciPENN, a multi-use deep learning approach that supports CITE-seq and scRNA-seq data integration, protein expression prediction for scRNA-seq, protein expression imputation for CITE-seq, quantification of prediction and imputation uncertainty, and cell type label transfer from CITE-seq to scRNA-seq. Comprehensive evaluations spanning multiple datasets demonstrate that sciPENN outperforms other current state-of-the-art methods.

Lakkis Justin, Schroeder Amelia, Su Kenong, Lee Michelle Y Y, Bashore Alexander C, Reilly Muredach P, Li Mingyao

2022-Nov

CITE-seq, deep learning, protein prediction, single-cell RNA-seq, single-cell multi-omics

General General

Identifying resilience strategies for disruption management in the healthcare supply chain during COVID-19 by digital innovations: A systematic literature review.

In Informatics in medicine unlocked

The worldwide spread of the COVID-19 disease has had a catastrophic effect on healthcare supply chains. The current manuscript systematically analyzes existing studies mitigating strategies for disruption management in the healthcare supply chain during COVID-19. Using a systematic approach, we recognized 35 related papers. Artificial intelligence (AI), block chain, big data analytics, and simulation are the most important technologies employed in supply chain management in healthcare. The findings reveal that the published research has concentrated mainly on generating resilience plans for the management of COVID-19 impacts. Furthermore, the vulnerability of healthcare supply chains and the necessity of establishing better resilience methods are emphasized in most of the research. However, the practical application of these emerging tools for managing disturbance and warranting resilience in the supply chain has been examined only rarely. This article provides directions for additional research, which can guide researchers to develop and conduct impressive studies related to the healthcare supply chain for different disasters.

Arji Goli, Ahmadi Hossein, Avazpoor Pejman, Hemmat Morteza

2023

COVID-19, Healthcare supply chain, Literature review, Pandemics, Supply chain management