Health Information Science and Systems最新文献

Purpose: Ultrasound image acquisition has the advantages of being low cost, rapid, and non-invasive, and it does not produce radiation. Currently, ultrasound is widely used in the diagnosis of liver tumors. However, owing to the complex presentation and diverse features of benign and malignant liver tumors, accurate diagnosis of liver tumors using ultrasound is difficult even for experienced radiologists. In recent years, artificial intelligence-assisted diagnosis has proven to provide effective support to radiologists. However, there is room for further improvement in the existing ultrasound artificial intelligence diagnostic model of liver tumor. First, the image diagnostic model may not fully consider relevant clinical data in the decision-making process. Second, owing to the difficulty in collecting biopsy pathology and physician-labeled ultrasound data of liver tumors, training datasets are usually small, and commonly used large neural networks tend to overfit on small datasets, which seriously affects the generalization of the model.

Methods: In this study, we propose a deep learning-assisted diagnosis model called USC-ENet, which integrates B-mode ultrasound features of liver tumors and clinical data of patients, and we design a small neural network specifically for small-scale medical images combined with an attention mechanism.

Results and conclusion: Real data from 542 patients with liver tumors (N = 2168 images) are used during model training and validation. Experiments show that USC-ENet can achieve a good classification effect (area under the curve = 0.956, sensitivity = 0.915, and specificity = 0.880) after small-scale data training, and it has certain interpretability, showing good potential for clinical adoption. In conclusion, our model provides not only a reliable second opinion for radiologists but also a reference for junior radiologists who lack clinical experience.

Purpose: Cardiotocography (CTG), which measures uterine contraction (UC) and fetal heart rate (FHR), is a crucial tool for assessing fetal health during pregnancy. However, traditional computerized cardiotocography (cCTG) approaches have non-negligible calibration errors in feature extraction and heavily rely on the expertise and prior experience to define diagnostic features from CTG or FHR signals. Although previous works have studied deep learning methods for extracting CTG or FHR features, these methods still neglect the clinical information of pregnant women.

Methods: In this paper, we proposed a multimodal deep learning architecture (MMDLA) for intelligent antepartum fetal monitoring that is capable of performing automatic CTG feature extraction, fusion with clinical data and classification. The multimodal feature fusion was achieved by concatenating high-level CTG features, which were extracted from preprocessed CTG signals via a convolution neural network (CNN) with six convolution layers and five fully connected layers, and the clinical data of pregnant women. Eventually, light gradient boosting machine (LGBM) was implemented as fetal status assessment classifier. The effectiveness of MMDLA was evaluated using a dataset of 16,355 cases, each of which includes FHR signal, UC signal and pertinent clinical data like maternal age and gestational age.

Results: With an accuracy of 90.77% and an area under the curve (AUC) value of 0.9201, the multimodal features performed admirably. The data imbalance issue was also effectively resolved by the LGBM classifier, with a normal-F1 value of 0.9376 and an abnormal-F1 value of 0.8223.

Conclusion: In summary, the proposed MMDLA is conducive to the realization of intelligent antepartum fetal monitoring.

Purpose: Telemedicine are experiencing an unprecedented boom globally since the beginning of the COVID-19 pandemic. As the most vulnerable groups amid COVID-19, the digital delivery of healthcare poses great challenges to the elderly population, caregiver, health service providers, and health policy makers. To bridge the service delivery gaps between the telemedicine demand side and supply side, explore evidence-based approach for integrated care, address challenges for aging policy, and build foundation for the development of data-driven and community-based telemedicine, our R&D team applied translational research to design and develop telemedicine "SMART" for enhancing elderly mental health wellbeing amid COVID-19. Our aim is to investigate the preparedness mechanisms of mental health disease including response, intervention, and connection these three healthcare delivery pipelines with the collection, consolidation, and synergy of heath parameters and social determinants, using data analytics approach to achieve Evidence-Based Medicine (EBM).

Methods: A mix of quantitative and qualitative research design for scientifically rigorous consultation and analysis was conducted from Jan 2020 to June 2021 in Hong Kong. An exploratory and descriptive qualitative design was used in this study. The data were collected through focus group discussions conducted from elderly and their caregivers living in 10 main districts of Hong Kong. Our research pilot tested "SMART" targeting for elderly with mental health improvement needs. Baseline questionnaire with 110 tele-medicine product users includes questions on demographic information, self-rated mental health digital adoption. The follow-up five focus group discussions with 57 users (elderly and their caregivers) further explore the social determinants of telemedicine transformation and help propose the integrated telemedicine paradigm shift framework establishment, development, and enhancement.

Results: Grounded on the baseline needs assessment and feedbacks collected, it is evident that multi-dimensional health information from the four various streams (community, clinic, home, remote) and customized digital health solutions are playing a key role in addressing elderly mental health digital service needs and bridging digital divide. The designed tele-medicine product lines up health service provider (supplier side) and elderly specific needs (demand side) with our three-level design, enables elderly and their families to follow and control their own health management and connect with the service provider, community of practice (CoP), and health policy makers.

Conclusion: It's beneficial to involve elderly and gerontechnology stakeholders as part of Community-Based Participatory Research (CBPR) before and throughout the developing and delivery phases an integrated and age-friendly digital intervention. The challenges in

Biomedical image segmentation plays a central role in quantitative analysis, clinical diagnosis, and medical intervention. In the light of the fully convolutional networks (FCN) and U-Net, deep convolutional networks (DNNs) have made significant contributions to biomedical image segmentation applications. In this paper, we propose three different multi-scale dense connections (MDC) for the encoder, the decoder of U-shaped architectures, and across them. Based on three dense connections, we propose a multi-scale densely connected U-Net (MDU-Net) for biomedical image segmentation. MDU-Net directly fuses the neighboring feature maps with different scales from both higher layers and lower layers to strengthen feature propagation in the current layer. Multi-scale dense connections, which contain shorter connections between layers close to the input and output, also make a much deeper U-Net possible. Besides, we introduce quantization to alleviate the potential overfitting in dense connections, and further improve the segmentation performance. We evaluate our proposed model on the MICCAI 2015 Gland Segmentation (GlaS) dataset. The three MDC improve U-Net performance by up to 1.8% on test A and 3.5% on test B in the MICCAI Gland dataset. Meanwhile, the MDU-Net with quantization obviously improves the segmentation performance of original U-Net.

Objective: The event-related potential (ERP) methods based on laboratory control scenes have been widely used to measure the level of mental workload during operational tasks. In this study, both task difficulty and test time were considered. Auditory probes (ignored task-irrelevant background sounds) were used to explore the changes in mental workload of unmanned aerial vehicle (UAV) operators during task execution and their ERP representations.

Approach: 51 students participated in a 10-day training and test of simulated quadrotor UAV. During the experiment, background sound was played to induce ERP according to the requirements of oddball paradigm, and the relationship between mental workload and the amplitudes of N200 and P300 in ERP was explored.

Main results: Our study shows that the mental workload during operational task training is multi-dimensional, and its changes are affected by bottom-up perception and top-down cognition. The N200 component of the ERP evoked by the auditory probe corresponds to the bottom-up perceptual part; while the P300 component corresponds to the top-down cognitive part, which is positively correlated with the improvement of skill level.

Significance: This paper describes the relationship between ERP induced by auditory probes and mental workload from the perspective of multi-resource theory and human information processing. This suggests that the auditory probe can be used to reveal the mental workload during the training of operational tasks, which not only provides a possible reference for measuring the mental workload, but also provides a possibility for identifying the development of the operator's skill level and evaluating the training effect.

Medical image segmentation is a challenging task due to the high variation in shape, size and position of infections or lesions in medical images. It is necessary to construct multi-scale representations to capture image contents from different scales. However, it is still challenging for U-Net with a simple skip connection to model the global multi-scale context. To overcome it, we proposed a dense skip-connection with cross co-attention in U-Net to solve the semantic gaps for an accurate automatic medical image segmentation. We name our method MCA-UNet, which enjoys two benefits: (1) it has a strong ability to model the multi-scale features, and (2) it jointly explores the spatial and channel attentions. The experimental results on the COVID-19 and IDRiD datasets suggest that our MCA-UNet produces more precise segmentation performance for the consolidation, ground-glass opacity (GGO), microaneurysms (MA) and hard exudates (EX). The source code of this work will be released via https://github.com/McGregorWwww/MCA-UNet/.

Purpose: In order to meet restrictions and difficulties in the development of hospital medical informatization and clinical databases in China, in this study, a disease-specific clinical database system (DSCDS) was designed and built. It provides support for the full utilization of real world medical big data in clinical research and medical services for specific diseases.

Methods: The development of DSCDS involved (1) requirements analysis on precision medicine, medical big data, and clinical research; (2) design schematics and basic architecture; (3) standard datasets of specific diseases consisting of common data elements (CDEs); (4) collection and aggregation of specific disease data scattered in various medical business systems of the hospital; (5) governance and quality improvement of specific disease data; (6) data storage and computing; and (7) design of data application modules.

Results: A DSCDS for liver cirrhosis was created in the gastrointestinal department of a 3A grade hospital in China and had more than nine data application modules. Based on this DSCDS, a series of clinical studies are being carried out, such as retrospective or prospective cohorts, prognostic studies using multimodal data, and follow-up studies.

Conclusion: The development of the DSCDS for liver cirrhosis in this paper provides experience and reference for the design and development of DSCDSs for other specific diseases in China; it can even expand to the development of DSCDSs in other countries if they have the demand for DSCDS and the same or better medical informatization foundation. DSCDS has more accurate, standard, comprehensive, multimodal and usable data of specific diseases than the general clinical database system and clinical data repository (CDR) and provides a credible data foundation for medical research, clinical decision-making and improving the medical service quality of specific diseases.

Supplementary information: The online version contains supplementary material available at 10.1007/s13755-023-00211-4.

3D medical image segmentation has an essential role in medical image analysis, while attention mechanism has improved the performance by a large margin. However, existing methods obtained the attention coefficient in a small receptive field, resulting in possible performance limitations. Radiologists usually scan all the slices first to have an overall idea of the target, and then analyze regions of interest in multiple 2D views in clinic practice. We simulate radiologists' recognition process and propose to exploit the 3D context information in a deeper manner for accurate 3D medical images segmentation. Due to the similarity of human body structure, medical images of different populations have highly similar shape and location information, so we use target region distillation to extract the common segmented region information. Particularly, we proposed two optimizations including Target Area Distillation and Section Attention. Target Area Distillation adds positions information to the original input to let the network has an initial attention of the target, while section attention performs attention extraction in three 2D sections thus with large range of receptive field. We compare our method against several popular networks in two public datasets including ImageCHD and COVID-19. Experimental results show that our proposed method improves the segmentation Dice score by 2-4% over the state-of-the-art methods. Our code has been released to the public (Anonymous link).

Wireless body area network (WBAN) is widely adopted in healthcare services, providing remote real-time and continuous healthcare monitoring. With the massive increase of detective sensor data, WBAN is largely restricted by limited storage and computation capacity, resulting in severely decreased efficiency and reliability. Mobile edge computing (MEC) technique can be combined with WBAN to resolve this issue. This paper studies the joint optimization problem of computational offloading and resource allocation (JCORA) in MEC for healthcare service scenarios. We formulate JCORA as a Markov decision process and propose a deep deterministic policy gradient-based WBAN offloading strategy (DDPG-WOS) to optimize time delay and energy consumption in interfered transmission channels. This scheme employs MEC to mitigate the computation pressure on a single WBAN and increase the transmission ability. Further, DDPG-WOS optimizes the offloading strategy-making process by considering the channel condition, transmission quality, computation ability and energy consumption. Simulation results verify the effectiveness of the proposed optimization schema in reducing energy consumption and computation latency and increasing the utility of WBAN compared to two competitive solutions.

Purpose: The early detection of organ failure mitigates the risk of post-intensive care syndrome and long-term functional impairment. The aim of this study is to predict organ failure in real-time for critical care patients based on a data-driven and knowledge-driven machine learning method (DKM) and provide explanations for the prediction by incorporating a medical knowledge graph.

Methods: The cohort of this study was a subset of the 4,386 adult Intensive Care Unit (ICU) patients from the MIMIC-III dataset collected between 2001 and 2012, and the primary outcome was the Delta Sequential Organ Failure Assessment (SOFA) score. A real-time Delta SOFA score prediction model was developed with two key components: an improved deep learning temporal convolutional network (S-TCN) and a graph-embedding feature extraction method based on a medical knowledge graph. Entities and relations related to organ failure were extracted from the Unified Medical Language System to build the medical knowledge graph, and patient data were mapped onto the graph to extract the embeddings. We measured the performance of our DKM approach with cross-validation to avoid the formation of biased assessments.

Results: An area under the receiver operating characteristic curve (AUC) of 0.973, a precision of 0.923, a NPV of 0.989, and an F1 score of 0.927 were achieved using the DKM approach, which significantly outperformed the baseline methods. Additionally, the performance remained stable following external validation on the eICU dataset, which consists of 2,816 admissions (AUC = 0.981, precision = 0.860, NPV = 0.984). Visualization of feature importance for the Delta SOFA score and their relationships on the basic clinical medical (BCM) knowledge graph provided a model explanation.

Conclusion: The use of an improved TCN model and a medical knowledge graph led to substantial improvement in prediction accuracy, providing generalizability and an independent explanation for organ failure prediction in critical care patients. These findings show the potential of incorporating prior domain knowledge into machine learning models to inform care and service planning.

Supplementary information: The online version of this article contains supplementary material available 10.1007/s13755-023-00210-5.