Search

Each year there are nearly 57 million deaths worldwide, with over 2.7 million in the United States. Timely, accurate and complete death reporting is critical for public health, especially during the COVID-19 pandemic, as institutions and government agencies rely on death reports to formulate responses to communicable diseases. Unfortunately, determining the causes of death is challenging even for experienced physicians. The novel coronavirus and its variants may further complicate the task, as physicians and experts are still investigating COVID-related complications. To assist physicians in accurately reporting causes of death, an advanced Artificial Intelligence (AI) approach is presented to determine a chronically ordered sequence of conditions that lead to death (named as the causal sequence of death), based on decedent's last hospital discharge record. The key design is to learn the causal relationship among clinical codes and to identify death-related conditions. There exist three challenges: different clinical coding systems, medical domain knowledge constraint, and data interoperability. First, we apply neural machine translation models with various attention mechanisms to generate sequences of causes of death. We use the BLEU (BiLingual Evaluation Understudy) score with three accuracy metrics to evaluate the quality of generated sequences. Second, we incorporate expert-verified medical domain knowledge as constraints when generating the causal sequences of death. Lastly, we develop a Fast Healthcare Interoperability Resources (FHIR) interface that demonstrates the usability of this work in clinical practice. Our results match the state-of-art reporting and can assist physicians and experts in public health crisis such as the COVID-19 pandemic.

Researchers seek help from deep learning methods to alleviate the enormous burden of reading radiological images by clinicians during the COVID-19 pandemic. However, clinicians are often reluctant to trust deep models due to their black-box characteristics. To automatically differentiate COVID-19 and community-acquired pneumonia from healthy lungs in radiographic imaging, we propose an explainable attention-transfer classification model based on the knowledge distillation network structure. The attention transfer direction always goes from the teacher network to the student network. Firstly, the teacher network extracts global features and concentrates on the infection regions to generate attention maps. It uses a deformable attention module to strengthen the response of infection regions and to suppress noise in irrelevant regions with an expanded reception field. Secondly, an image fusion module combines attention knowledge transferred from teacher network to student network with the essential information in original input. While the teacher network focuses on global features, the student branch focuses on irregularly shaped lesion regions to learn discriminative features. Lastly, we conduct extensive experiments on public chest X-ray and CT datasets to demonstrate the explainability of the proposed architecture in diagnosing COVID-19.

Patient and health provider interaction via text messaging (TM) has become an accepted form of communication, often favored by adolescents and young adults. While integration of TM in disease management has aided health interventions and behavior modifications, broader adoption is hindered by expense, fixed reporting schedules, and monotonic communication. A low-cost, flexible TM reporting system (REMOTES) was developed using inexpensive cloud-based services with features of two-way communication, personalized reporting scheduling, and scalable and secured data storage. REMOTES is a template-based reporting tool adaptable to a wide-range of complexity in response formats. In a pilot study, 27 adolescents with sickle cell disease participated to assess feasibility of REMOTES in both inpatient and outpatient settings. Subject compliance with at least one daily self-report pain query was 94.9% (112/118) during inpatient and 91.1% (327/359) during outpatient, with an overall accuracy of 99.2% (970/978). With use of a more complex 8-item questionnaire, 30% (7/21) inpatient and 66.6% (36/54) outpatient responses were reported with 98.1% (51/52) reporting accuracy. All participants expressed high pre-trial expectation (88%) and post-trial satisfaction (89%). The study suggests that cloud-based text messaging is feasible and an easy-of-use solution for low-cost and personalized patient engagement.

The diversity and number of parameters monitored in an intensive care unit (ICU) make the resulting databases highly susceptible to quality issues, such as missing information and erroneous data entry, which adversely affect the downstream processing and predictive modeling. Missing data interpolation and imputation techniques, such as multiple imputation, expectation maximization, and hot-deck imputation techniques do not account for the type of missing data, which can lead to bias. In our study, we first model the missing data as three types: "neglectable" also known as a.k.a "missing completely at random," "recoverable" a.k.a. "missing at random," and "not easily recoverable" a.k.a. "missing not at random." We then design imputation techniques for each type of missing data. We use a publicly available database (MIMIC II) to demonstrate how these imputations perform with random forests for prediction. Our results indicate that these novel imputation techniques outperformed standard mean filling techniques and expectation maximization with a statistical significance p ≤ 0.01 in predicting ICU mortality.

One pressing need in the area of public health is timely, accurate, and complete reporting of deaths and the diseases or conditions leading up to them. Fast Healthcare Interoperability Resources (FHIR) is a new HL7 interoperability standard for electronic health record, while Sustainable Medical Applications and Reusable Technologies (SMART)-on-FHIR enables third-party app development that can work "out of the box." This paper demonstrates the feasibility of developing SMART-on-FHIR applications that enables medical professionals to perform timely and accurate death reporting within multiple different USA State jurisdictions. We explored how the information on a standard certificate of death can be mapped to resources defined in the FHIR standard Draft Standard for Trial Use Version 2 and common profiles. We also demonstrated analytics for potentially improving the accuracy and completeness of mortality reporting data.

Objective: Rapid advances of high-throughput technologies and wide adoption of electronic health records (EHRs) have led to fast accumulation of -omic and EHR data. These voluminous complex data contain abundant information for precision medicine, and big data analytics can extract such knowledge to improve the quality of healthcare., Methods: In this paper, we present -omic and EHR data characteristics, associated challenges, and data analytics including data preprocessing, mining, and modeling., Results: To demonstrate how big data analytics enables precision medicine, we provide two case studies, including identifying disease biomarkers from multi-omic data and incorporating -omic information into EHR., Conclusion: Big data analytics is able to address -omic and EHR data challenges for paradigm shift toward precision medicine., Significance: Big data analytics makes sense of -omic and EHR data to improve healthcare outcome. It has long lasting societal impact.

Late diagnosis is one of the reasons that head and neck squamous cell carcinoma (HNSCC) patients experience relative five-year survival rates ranging from 40%-66%. The molecular-level differences between early and advanced stage HNSCC may provide insight into therapeutic targets and strategies. Previous bioinformatics studies have shown mixed or limited results in identifying gene and protein markers and in developing models for discriminating between early and advanced stage HNSCC. Thus, we have investigated models for HNSCC stage prediction using RNAseq and reverse phase protein array data from The Cancer Genome Atlas and The Cancer Proteome Atlas. We systematically assessed individual and ensemble binary classifiers, using filter and wrapper feature selection methods, to develop several well-performing models. In particular, integrated models harnessing both data types consistently resulted in better performance. This study identifies informative protein and gene feature sets which may increase understanding of HNSCC progression.

Objective: High prevalence of diabetes mellitus (DM) along with the poor health outcomes and the escalated costs of treatment and care poses the need to focus on prevention, early detection and improved management of the disease. The aim of this paper is to present and discuss the latest accomplishments in sensors for glucose and lifestyle monitoring along with clinical decision support systems (CDSSs) facilitating self-disease management and supporting healthcare professionals in decision making., Methods: A critical literature review analysis is conducted focusing on advances in: 1) sensors for physiological and lifestyle monitoring, 2) models and molecular biomarkers for predicting the onset and assessing the progress of DM, and 3) modeling and control methods for regulating glucose levels., Results: Glucose and lifestyle sensing technologies are continuously evolving with current research focusing on the development of noninvasive sensors for accurate glucose monitoring. A wide range of modeling, classification, clustering, and control approaches have been deployed for the development of the CDSS for diabetes management. Sophisticated multiscale, multilevel modeling frameworks taking into account information from behavioral down to molecular level are necessary to reveal correlations and patterns indicating the onset and evolution of DM., Conclusion: Integration of data originating from sensor-based systems and electronic health records combined with smart data analytics methods and powerful user centered approaches enable the shift toward preventive, predictive, personalized, and participatory diabetes care., Significance: The potential of sensing and predictive modeling approaches toward improving diabetes management is highlighted and related challenges are identified.

This paper summarizes the panel discussion at the IEEE Engineering in Medicine and Biology Point-of-Care Healthcare Technology Conference (POCHT 2013) held in Bangalore India from Jan 16-18, 2013. Modern medicine has witnessed interdisciplinary technology innovations in healthcare with a continuous growth in life expectancy across the globe. However, there is also a growing global concern on the affordability of rapidly rising healthcare costs. To provide quality healthcare at reasonable costs, there has to be a convergence of preventive, personalized, and precision medicine with the help of technology innovations across the entire spectrum of point-of-care (POC) to critical care at hospitals. The first IEEE EMBS Special Topic POCHT conference held in Bangalore, India provided an international forum with clinicians, healthcare providers, industry experts, innovators, researchers, and students to define clinical needs and technology solutions toward commercialization and translation to clinical applications across different environments and infrastructures. This paper presents a summary of discussions that took place during the keynote presentations, panel discussions, and breakout sessions on needs, challenges, and technology innovations in POC technologies toward improving global healthcare. Also presented is an overview of challenges and trends in developing and developed economies with respect to priority clinical needs, technology innovations in medical devices, translational engineering, information and communication technologies, infrastructure support, and patient and clinician acceptance of POC healthcare technologies.

Researchers have developed computer-aided decision support systems for translational medicine that aim to objectively and efficiently diagnose cancer using histopathological images. However, the performance of such systems is confounded by nonbiological experimental variations or "batch effects" that can commonly occur in histopathological data, especially when images are acquired using different imaging devices and patient samples. This is even more problematic in large-scale studies in which cross-laboratory sharing of large volumes of data is necessary. Batch effects can change quantitative morphological image features and decrease the prediction performance. Using four batches of renal tumor images, we compare one image-level and five feature-level batch effect removal methods. Principal component variation analysis shows that batch is a large source of variance in image features. Results show that feature-level normalization methods reduce batch-contributed variance to almost zero. Moreover, feature-level normalization, especially ComBatN, improves cross-batch and combined-batch prediction performance. Compared to no normalization, ComBatN improves performance in 83% and 90% of cross-batch and combined-batch prediction models, respectively.

We have developed omniBiomarker, a web-based application that uses knowledge from the NCI Cancer Gene Index to guide the selection of biologically relevant algorithms for identifying biomarkers. Biomarker identification from high-throughput genomic expression data is difficult because of data properties (i.e., small-sample size compared to large-feature size) as well as the large number of available feature selection algorithms. Thus, it is unclear which algorithm should be used for a particular dataset. These factors lead to instability in biomarker identification and affect the reproducibility of results. We introduce a method for computing the biological relevance of feature selection algorithms using an externally validated knowledge base of manually curated cancer biomarkers. Results suggest that knowledge-driven biomarker identification can improve microarray-based clinical prediction performance. omniBiomarker can be accessed at http://omnibiomarker.bme.gatech.edu/.

One way to gain a more comprehensive picture of the complex function of a cell is to study the transcriptome. A promising technology for studying the transcriptome is RNA sequencing, an application of which is to quantify elements in the transcriptome and to link quantitative observations to biology. Although numerous quantification algorithms are publicly available, no method of systematically assessing these algorithms has been developed. To meet the need for such an assessment, we present an approach that includes (1) simulated and real datasets, (2) three alignment strategies, and (3) six quantification algorithms. Examining the normalized root-mean-square error, the percentage error of the coefficient of variation, and the distribution of the coefficient of variation, we found that quantification algorithms with the input of sequence alignment reported in the transcriptomic coordinate usually performed better in terms of the multiple metrics proposed in this study.

This paper summarizes the discussions held during the First IEEE Life Sciences Grand Challenges Conference, held on October 4-5, 2012, at the National Academy of Sciences, Washington, DC, and the grand challenges identified by the conference participants. Despite tremendous efforts to develop the knowledge and ability that are essential in addressing biomedical and health problems using engineering methodologies, the optimization of this approach toward engineering the life sciences and healthcare remains a grand challenge. The conference was aimed at high-level discussions by participants representing various sectors, including academia, government, and industry. Grand challenges were identified by the conference participants in five areas including engineering the brain and nervous system; engineering the cardiovascular system; engineering of cancer diagnostics, therapeutics, and prevention; translation of discoveries to clinical applications; and education and training. A number of these challenges are identified and summarized in this paper.

Accurate quantification of gene or isoform expression with RNA-Seq depends on complete knowledge of the transcriptome. Because a complete genomic annotation does not yet exist, novel isoform discovery is an important component of the RNA-Seq quantification process. Thus, a typical RNA-Seq pipeline includes a transcriptome mapping step to quantify known genes and isoforms, and a reference genome mapping step to discover new genes and isoforms. Several tools implement this approach, but are limited in that they force the use of a single mapping algorithm at both the transcriptome and reference genome mapping stages. The choice of mapping algorithm could affect quantification accuracy on a per-dataset basis. Thus, we describe a method that enables the merging of transcriptome and reference genome mapping stages provided that they conform to the standard SAM/BAM format. This procedure could potentially improve the accuracy of gene or isoform quantification by increasing flexibility when selecting RNA-Seq data analysis pipelines. We demonstrate an example of a flexible RNA-Seq pipeline by assessing its potential for novel isoform discovery and by validating its quantification performance using qRT-PCR.