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

Surgery Surgery

Multi-omics and immune cells' profiling of COVID-19 patients for ICU admission prediction: in silico analysis and an integrated machine learning-based approach in the framework of Predictive, Preventive, and Personalized Medicine.

In The EPMA journal

BACKGROUND : Intensive care unit admission (ICUA) triage has been urgent need for solving the shortage of ICU beds, during the coronavirus disease 2019 (COVID-19) surge. In silico analysis and integrated machine learning (ML) approach, based on multi-omics and immune cells (ICs) profiling, might provide solutions for this issue in the framework of predictive, preventive, and personalized medicine (PPPM).

METHODS : Multi-omics was used to screen the synchronous differentially expressed protein-coding genes (SDEpcGs), and an integrated ML approach to develop and validate a nomogram for prediction of ICUA. Finally, the independent risk factor (IRF) with ICs profiling of the ICUA was identified.

RESULTS : Colony-stimulating factor 1 receptor (CSF1R) and peptidase inhibitor 16 (PI16) were identified as SDEpcGs, and each fold change (FCij) of CSF1R and PI16 was selected to develop and validate a nomogram to predict ICUA. The area under curve (AUC) of the nomogram was 0.872 (95% confidence interval (CI): 0.707 to 0.950) on the training set, and 0.822 (95% CI: 0.659 to 0.917) on the testing set. CSF1R was identified as an IRF of ICUA, expressed in and positively correlated with monocytes which had a lower fraction in COVID-19 ICU patients.

CONCLUSION : The nomogram and monocytes could provide added value to ICUA prediction and targeted prevention, which are cost-effective platform for personalized medicine of COVID-19 patients. The log2fold change (log2FC) of the fraction of monocytes could be monitored simply and economically in primary care, and the nomogram offered an accurate prediction for secondary care in the framework of PPPM.

SUPPLEMENTARY INFORMATION : The online version contains supplementary material available at 10.1007/s13167-023-00317-5.

Zhu Kun, Chen Zhonghua, Xiao Yi, Lai Dengming, Wang Xiaofeng, Fang Xiangming, Shu Qiang

2023-Feb-21

COVID-19, CSF1R, Immune cells, Machine learning, Monocytes, Nomogram, PI16, Predictive Preventive Personalized medicine (PPPM / 3PM), Predictive model, Triage

General General

A new Apache Spark-based framework for big data streaming forecasting in IoT networks.

In The Journal of supercomputing

Analyzing time-dependent data acquired in a continuous flow is a major challenge for various fields, such as big data and machine learning. Being able to analyze a large volume of data from various sources, such as sensors, networks, and the internet, is essential for improving the efficiency of our society's production processes. Additionally, this vast amount of data is collected dynamically in a continuous stream. The goal of this research is to provide a comprehensive framework for forecasting big data streams from Internet of Things networks and serve as a guide for designing and deploying other third-party solutions. Hence, a new framework for time series forecasting in a big data streaming scenario, using data collected from Internet of Things networks, is presented. This framework comprises of five main modules: Internet of Things network design and deployment, big data streaming architecture, stream data modeling method, big data forecasting method, and a comprehensive real-world application scenario, consisting of a physical Internet of Things network feeding the big data streaming architecture, being the linear regression the algorithm used for illustrative purposes. Comparison with other frameworks reveals that this is the first framework that incorporates and integrates all the aforementioned modules.

Fernández-Gómez Antonio M, Gutiérrez-Avilés David, Troncoso Alicia, Martínez-Álvarez Francisco

2023-Feb-21

Big data, IoT, Streaming analysis, Time series

General General

Artificial Intelligence and Cardiovascular Risk Prediction: All That Glitters is not Gold.

In European cardiology

Artificial intelligence (AI) is a broad term referring to any automated systems that need 'intelligence' to carry out specific tasks. During the last decade, AI-based techniques have been gaining popularity in a vast range of biomedical fields, including the cardiovascular setting. Indeed, the dissemination of cardiovascular risk factors and the better prognosis of patients experiencing cardiovascular events resulted in an increase in the prevalence of cardiovascular disease (CVD), eliciting the need for precise identification of patients at increased risk for development and progression of CVD. AI-based predictive models may overcome some of the limitations that hinder the performance of classic regression models. Nonetheless, the successful application of AI in this field requires knowledge of the potential pitfalls of the AI techniques, to guarantee their safe and effective use in daily clinical practice. The aim of the present review is to summarise the pros and cons of different AI methods and their potential application in the cardiovascular field, with a focus on the development of predictive models and risk assessment tools.

Chiarito Mauro, Luceri Luca, Oliva Angelo, Stefanini Giulio, Condorelli Gianluigi

2022-Feb

Artificial intelligence, cardiovascular disease, machine learning, risk prediction

General General

Pancancer survival prediction using a deep learning architecture with multimodal representation and integration.

In Bioinformatics advances

MOTIVATION : Use of multi-omics data carrying comprehensive signals about the disease is strongly desirable for understanding and predicting disease progression, cancer particularly as a serious disease with a high mortality rate. However, recent methods currently fail to effectively utilize the multi-omics data for cancer survival prediction and thus significantly limiting the accuracy of survival prediction using omics data.

RESULTS : In this work, we constructed a deep learning model with multimodal representation and integration to predict the survival of patients using multi-omics data. We first developed an unsupervised learning part to extract high-level feature representations from omics data of different modalities. Then, we used an attention-based method to integrate feature representations, produced by the unsupervised learning part, into a single compact vector and finally we fed the vector into fully connected layers for survival prediction. We used multimodal data to train the model and predict pancancer survival, and the results show that using multimodal data can lead to higher prediction accuracy compared to using single modal data. Furthermore, we used the concordance index and the 5-fold cross-validation method for comparing our proposed method with current state-of-the-art methods and our results show that our model achieves better performance on the majority of cancer types in our testing datasets.

AVAILABILITY AND IMPLEMENTATION : https://github.com/ZhangqiJiang07/MultimodalSurvivalPrediction.

SUPPLEMENTARY INFORMATION : Supplementary data are available at Bioinformatics online.

Fan Ziling, Jiang Zhangqi, Liang Hengyu, Han Chao

2023

General General

Interpretatively automated identification of circulating tumor cells from human peripheral blood with high performance.

In Frontiers in bioengineering and biotechnology

The detection and analysis of circulating tumor cells (CTCs) would be of aid in a precise cancer diagnosis and an efficient prognosis assessment. However, traditional methods that rely heavily on the isolation of CTCs based on their physical or biological features suffer from intensive labor, thus being unsuitable for rapid detection. Furthermore, currently available intelligent methods are short of interpretability, which creates a lot of uncertainty during diagnosis. Therefore, we propose here an automated method that takes advantage of bright-field microscopic images with high resolution, so as to take an insight into cell patterns. Specifically, the precise identification of CTCs was achieved by using an optimized single-shot multi-box detector (SSD)-based neural network with integrated attention mechanism and feature fusion modules. Compared to the conventional SSD system, our method exhibited a superior detection performance with the recall rate of 92.2%, and the maximum average precision (AP) value of 97.9%. To note, the optimal SSD-based neural network was combined with advanced visualization technology, i.e., the gradient-weighted class activation mapping (Grad-CAM) for model interpretation, and the t-distributed stochastic neighbor embedding (T-SNE) for data visualization. Our work demonstrates for the first time the outstanding performance of SSD-based neural network for CTCs identification in human peripheral blood environment, showing great potential for the early detection and continuous monitoring of cancer progression.

Li Xiaolei, Chen Mingcan, Xu Jingjing, Wu Dihang, Ye Mengxue, Wang Chi, Liu Wanyu

2023

circulating tumor cells, deep learning, interpretative analysis, precise identification, single-shot multibox detector

General General

Development of an automated biomaterial platform to study mosquito feeding behavior.

In Frontiers in bioengineering and biotechnology

Mosquitoes carry a number of deadly pathogens that are transmitted while feeding on blood through the skin, and studying mosquito feeding behavior could elucidate countermeasures to mitigate biting. Although this type of research has existed for decades, there has yet to be a compelling example of a controlled environment to test the impact of multiple variables on mosquito feeding behavior. In this study, we leveraged uniformly bioprinted vascularized skin mimics to create a mosquito feeding platform with independently tunable feeding sites. Our platform allows us to observe mosquito feeding behavior and collect video data for 30-45 min. We maximized throughput by developing a highly accurate computer vision model (mean average precision: 92.5%) that automatically processes videos and increases measurement objectivity. This model enables assessment of critical factors such as feeding and activity around feeding sites, and we used it to evaluate the repellent effect of DEET and oil of lemon eucalyptus-based repellents. We validated that both repellents effectively repel mosquitoes in laboratory settings (0% feeding in experimental groups, 13.8% feeding in control group, p < 0.0001), suggesting our platform's use as a repellent screening assay in the future. The platform is scalable, compact, and reduces dependence on vertebrate hosts in mosquito research.

Janson Kevin D, Carter Brendan H, Jameson Samuel B, de Verges Jane E, Dalliance Erika S, Royse Madison K, Kim Paul, Wesson Dawn M, Veiseh Omid

2023

3D printing, biofabrication, machine learning, mosquito repellent, mosquito-borne diseases, object detection