Development and quantitative assessment of deep learning-based image enhancement for optical coherence tomography.

PURPOSE: To develop a deep learning-based framework to improve the image quality of optical coherence tomography (OCT) and evaluate its image enhancement effect with the traditional image averaging method from a clinical perspective. METHODS: 359 normal eyes and 456 eyes with various retinal conditions were included. A deep learning framework with high-resolution representation was developed to achieve image quality enhancement for OCT images. The quantitative comparisons, including expert subjective scores from ophthalmologists and three objective metrics of image quality (structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR) and contrast-to-noise ratio (CNR)), were performed between deep learning method and traditional image averaging. RESULTS: With the increase of frame count from 1 to 20, our deep learning method always obtained higher SSIM and PSNR values than the image averaging method while importing the same number of frames. When we selected 5 frames as inputs, the local objective assessment with CNR illustrated that the deep learning method had more obvious tissue contrast enhancement than averaging method. The subjective scores of image quality were all highest in our deep learning method, both for normal retinal structure and various retinal lesions. All the objective and subjective indicators had significant statistical differences (P < 0.05). CONCLUSION: Compared to traditional image averaging methods, our proposed deep learning enhancement framework can achieve a reasonable trade-off between image quality and scanning times, reducing the number of repeated scans.

Automated bone mineral density prediction and fracture risk assessment using plain radiographs via deep learning.

Dual-energy X-ray absorptiometry (DXA) is underutilized to measure bone mineral density (BMD) and evaluate fracture risk. We present an automated tool to identify fractures, predict BMD, and evaluate fracture risk using plain radiographs. The tool performance is evaluated on 5164 and 18175 patients with pelvis/lumbar spine radiographs and Hologic DXA. The model is well calibrated with minimal bias in the hip (slope = 0.982, calibration-in-the-large = -0.003) and the lumbar spine BMD (slope = 0.978, calibration-in-the-large = 0.003). The area under the precision-recall curve and accuracy are 0.89 and 91.7% for hip osteoporosis, 0.89 and 86.2% for spine osteoporosis, 0.83 and 95.0% for high 10-year major fracture risk, and 0.96 and 90.0% for high hip fracture risk. The tool classifies 5206 (84.8%) patients with 95% positive or negative predictive value for osteoporosis, compared to 3008 DXA conducted at the same study period. This automated tool may help identify high-risk patients for osteoporosis.

Construction of a Non-Mutually Exclusive Decision Tree for Medication Recommendation of Chronic Heart Failure.

Objective: Although guidelines have recommended standardized drug treatment for heart failure (HF), there are still many challenges in making the correct clinical decisions due to the complicated clinical situations of HF patients. Each patient would satisfy several recommendations, meaning the decision tree of HF treatment should be nonmutually exclusive, and the same patient would be allocated to several leaf nodes in the decision tree. In the current study, we aim to propose a way to ensemble a nonmutually exclusive decision tree for recommendation system for complicated diseases, such as HF. Methods: The nonmutually exclusive decision tree was constructed via knowledge rules summarized from the HF clinical guidelines. Then similar patients were defined as those who followed the same pattern of leaf node allocation according to the decision tree. The frequent medication patterns for each similar patient were mined using the Apriori algorithms, and we also carried out the outcome prognosis analyses to show the capability for the evidence-based medication recommendations of our nonmutually exclusive decision tree. Results: Based on a large database that included 29,689 patients with 84,705 admissions, we tested the framework for HF treatment recommendation. In the constructed decision tree, the HF treatment recommendations were grouped into two independent parts. The first part was recommendations for new cases, and the second part was recommendations when patients had different historical medication. There are 14 leaf nodes in our decision tree, and most of the leaf nodes had a guideline adherence of around 90%. We reported the top 10 popular similar patients, which accounted for 32.84% of the whole population. In addition, the multiple outcome prognosis analyses were carried out to assess the medications for one of the subgroups of similar patients. Our results showed even for the subgroup of the same similar patients that no one medication pattern would benefit all outcomes. Conclusion: In the present study, the methodology to construct a nonmutually exclusive decision tree for medication recommendations for HF and its application in CDSS was proposed. Our framework is universal for most diseases and could be generally applied in developing the CDSS for treatment.

Construction of Guideline-Based Decision Tree for Medication Recommendation.

Clinical decision support system (CDSS) plays an essential role nowadays and CDSS for treatment provides clinicians with the clinical evidence of candidate prescriptions to assist them in making patient-specific decisions. Therefore, it is essential to find a partition of patients such that patients with similar clinical conditions are grouped together and the preferred prescriptions for different groups are diverged. A comprehensive clinical guideline often provides information of patient partition. However, for most diseases, the guideline is not so detailed that only limited circumstances are covered. This makes it challenging to group patients properly. Here we proposed an approach that combines clinical guidelines with medical data to construct a nested decision tree for patient partitioning and treatment recommendation. Compared with pure data-driven decision tree, the recommendations generated by our model have better guideline adherence and interpretability. The approach was successfully applied in a real-world case study of patients with hyperthyroidism.

Analysis of the Health Information Needs of Diabetics in China.

According to the latest statistics of the China National Health Protection Commission, the prevalence of adult diabetics in China has reached 11.6%, and the number of patients has exceeded 114 million. Understanding the needs of diabetics and what kind of problems they are anxious about are crucial for doctors, hospitals and other health care providers, which can be used to ameliorate patient education services and help patients to improve their disease management skills. Hence we have conducted a study to analyze the questions about diabetes collected from a Chinese health website; the number of which is 151,589. We have divided these questions into 9 categories using a convolutional neural network. The shocking results showed that the questions about prevention only account for 1.23%. And a Chinese patented drug Xiaoke pill, the main component of which is glibenclamide, ranks fourth among the drugs the user cares most about due to the cheap price. However, patients know very little about its side effects.

Precision Health Economics and Outcomes Research to Support Precision Medicine: Big Data Meets Patient Heterogeneity on the Road to Value.

The "big data" era represents an exciting opportunity to utilize powerful new sources of information to reduce clinical and health economic uncertainty on an individual patient level. In turn, health economic outcomes research (HEOR) practices will need to evolve to accommodate individual patient-level HEOR analyses. We propose the concept of "precision HEOR", which utilizes a combination of costs and outcomes derived from big data to inform healthcare decision-making that is tailored to highly specific patient clusters or individuals. To explore this concept, we discuss the current and future roles of HEOR in health sector decision-making, big data and predictive analytics, and several key HEOR contexts in which big data and predictive analytics might transform traditional HEOR into precision HEOR. The guidance document addresses issues related to the transition from traditional to precision HEOR practices, the evaluation of patient similarity analysis and its appropriateness for precision HEOR analysis, and future challenges to precision HEOR adoption. Precision HEOR should make precision medicine more realizable by aiding and adapting healthcare resource allocation. The combined hopes for precision medicine and precision HEOR are that individual patients receive the best possible medical care while overall healthcare costs remain manageable or become more cost-efficient.

Feature++: Automatic Feature Construction for Clinical Data Analysis.

With the rapid growth of clinical data and knowledge, feature construction for clinical analysis becomes increasingly important and challenging. Given a clinical dataset with up to hundreds or thousands of columns, the traditional manual feature construction process is usually too labour intensive to generate a full spectrum of features with potential values. As a result, advanced large-scale data analysis technologies, such as feature selection for predictive modelling, cannot be fully utilized for clinical data analysis. In this paper, we propose an automatic feature construction framework for clinical data analysis, namely, Feature++. It leverages available public knowledge to understand the semantics of the clinical data, and is able to integrate external data sources to automatically construct new features based on predefined rules and clinical knowledge. We demonstrate the effectiveness of Feature++ in a typical predictive modelling use case with a public clinical dataset, and the results suggest that the proposed approach is able to fulfil typical feature construction tasks with minimal dataset specific configurations, so that more accurate models can be obtained from various clinical datasets in a more efficient way.

Automatic variance analysis of multistage care pathways.

A care pathway (CP) is a standardized process that consists of multiple care stages, clinical activities and their relations, aimed at ensuring and enhancing the quality of care. However, actual care may deviate from the planned CP, and analysis of these deviations can help clinicians refine the CP and reduce medical errors. In this paper, we propose a CP variance analysis method to automatically identify the deviations between actual patient traces in electronic medical records (EMR) and a multistage CP. As the care stage information is usually unavailable in EMR, we first align every trace with the CP using a hidden Markov model. From the aligned traces, we report three types of deviations for every care stage: additional activities, absent activities and violated constraints, which are identified by using the techniques of temporal logic and binomial tests. The method has been applied to a CP for the management of congestive heart failure and real world EMR, providing meaningful evidence for the further improvement of care quality.

A framework for evaluating sequential clinical interventions with disease progression.

This work proposes to leverage an advanced modeling technique, namely Markov Decision Process, to evaluate sequential clinical interventions in disease management. We have demonstrated our evaluation framework on a diabetes case study over two real data sets, and discovered valuable clinical insights towards better interventions during disease progression.

An efficacy driven approach for medication recommendation in type 2 diabetes treatment using data mining techniques.

We demonstrate how data mining techniques can help recommend effective medications when physicians need to control the glucose level of patients with type 2 diabetes. We first identify the factors that may affect physicians' medication decisions and then develop a patient-similarity based approach to automatically recommend medications for a patient with the specific condition so that his blood glucose level (measured by HbA1C value) can be well controlled. The approach is validated through experiments on real data sets and compared with the recommendations by following a clinical guideline.

Beyond regional health information exchange in China: a practical and industrial-strength approach.

In this paper, we present the design and implementation of a regional health information system that reconciles patient clinical data from heterogeneous Point of Services(POS) applications and supports complicated clinical queries. We propose to design a simple XML format for the representation of clinical documents and a messaging-based protocol for exchanging of clinical documents to facilitate the adoption of international standards such as CDA and IHE XDS profile for local application vendors. We also propose to use a hybrid relational-XML database for the storage of CDA documents that leverages both relational and XML optimization techniques to improve the performance of flexible clinical queries. This system has been deployed in a pilot phase to a regional health information organization operated by a top hospital in Beijing, China.