TERMS: textual emotion recognition in multidimensional space.

Microblogs generate a vast amount of data in which users express their emotions regarding almost all aspects of everyday life. Capturing affective content from these context-dependent and subjective texts is a challenging task. We propose an intelligent probabilistic model for textual emotion recognition in multidimensional space (TERMS) that captures the subjective emotional boundaries and contextual information embedded in a text for robust emotion recognition. It is implausible with discrete label assignment;therefore, the model employs a soft assignment by mapping varying emotional perceptions in a multidimensional space and generates them as distributions via the Gaussian mixture model (GMM). To strengthen emotion distributions, TERMS integrates a probabilistic emotion classifier that captures the contextual and linguistic information from texts. The integration of these aspects, the context-aware emotion classifier and the learned GMM parameters provide a complete coverage for accurate emotion recognition. The large-scale experimentation shows that compared to baseline and state-of-the-art models, TERMS achieved better performance in terms of distinguishability, prediction, and classification performance. In addition, TERMS provide insights on emotion classes, the annotation patterns, and the models application in different scenarios.

A machine learning approach for predicting urine output after fluid administration.

BACKGROUND AND OBJECTIVE: To develop a machine learning model to predict urine output (UO) in sepsis patients after fluid resuscitation. METHODS: We identified sepsis patients in the Multiparameter Intelligent Monitoring in Intensive Care-III v1.4 database according to the Sepsis-3 criteria. We focused on two outcomes: whether the UO decreased after fluid administration and whether oliguria (defined as UO less than the threshold of 0.5 mL/kg/h) developed. A gradient tree-based machine learning model implemented with an eXtreme Gradient Boosting algorithm was used to integrate relevant physiological parameters for predicting the aforementioned outcomes. A confusion matrix was computed. RESULTS: A total of 232,929 events in 19,275 patients were included. Using decreased UO as the outcome measure, the optimal model achieved an area under the curve (AUC) of 0.86; for predicting oliguria, most models achieved an AUC greater than 0.86, and the highest sensitivity was 92.2% when the model was applied to patients with baseline oliguria. CONCLUSIONS: Machine learning could help clinicians evaluate fluid status in sepsis patients after fluid administration, thus preventing fluid overload-related complications.

Automation of the kidney function prediction and classification through ultrasound-based kidney imaging using deep learning.

Prediction of kidney function and chronic kidney disease (CKD) through kidney ultrasound imaging has long been considered desirable in clinical practice because of its safety, convenience, and affordability. However, this highly desirable approach is beyond the capability of human vision. We developed a deep learning approach for automatically determining the estimated glomerular filtration rate (eGFR) and CKD status. We exploited the transfer learning technique, integrating the powerful ResNet model pretrained on an ImageNet dataset in our neural network architecture, to predict kidney function based on 4,505 kidney ultrasound images labeled using eGFRs derived from serum creatinine concentrations. To further extract the information from ultrasound images, we leveraged kidney length annotations to remove the peripheral region of the kidneys and applied various data augmentation schemes to produce additional data with variations. Bootstrap aggregation was also applied to avoid overfitting and improve the model's generalization. Moreover, the kidney function features obtained by our deep neural network were used to identify the CKD status defined by an eGFR of <60 ml/min/1.73 m2. A Pearson correlation coefficient of 0.741 indicated the strong relationship between artificial intelligence (AI)- and creatinine-based GFR estimations. Overall CKD status classification accuracy of our model was 85.6% -higher than that of experienced nephrologists (60.3%-80.1%). Our model is the first fundamental step toward realizing the potential of transforming kidney ultrasound imaging into an effective, real-time, distant screening tool. AI-GFR estimation offers the possibility of noninvasive assessment of kidney function, a key goal of AI-powered functional automation in clinical practice.

Synthesizing electronic health records using improved generative adversarial networks.

Objective: The aim of this study was to generate synthetic electronic health records (EHRs). The generated EHR data will be more realistic than those generated using the existing medical Generative Adversarial Network (medGAN) method. Materials and Methods: We modified medGAN to obtain two synthetic data generation models-designated as medical Wasserstein GAN with gradient penalty (medWGAN) and medical boundary-seeking GAN (medBGAN)-and compared the results obtained using the three models. We used 2 databases: MIMIC-III and National Health Insurance Research Database (NHIRD), Taiwan. First, we trained the models and generated synthetic EHRs by using these three 3 models. We then analyzed and compared the models' performance by using a few statistical methods (Kolmogorov-Smirnov test, dimension-wise probability for binary data, and dimension-wise average count for count data) and 2 machine learning tasks (association rule mining and prediction). Results: We conducted a comprehensive analysis and found our models were adequately efficient for generating synthetic EHR data. The proposed models outperformed medGAN in all cases, and among the 3 models, boundary-seeking GAN (medBGAN) performed the best. Discussion: To generate realistic synthetic EHR data, the proposed models will be effective in the medical industry and related research from the viewpoint of providing better services. Moreover, they will eliminate barriers including limited access to EHR data and thus accelerate research on medical informatics. Conclusion: The proposed models can adequately learn the data distribution of real EHRs and efficiently generate realistic synthetic EHRs. The results show the superiority of our models over the existing model.

Predicting financial trouble using call data-On social capital, phone logs, and financial trouble.

An ability to understand and predict financial wellbeing for individuals is of interest to economists, policy designers, financial institutions, and the individuals themselves. According to the Nilson reports, there were more than 3 billion credit cards in use in 2013, accounting for purchases exceeding US$ 2.2 trillion, and according to the Federal Reserve report, 39% of American households were carrying credit card debt from month to month. Prior literature has connected individual financial wellbeing with social capital. However, as yet, there is limited empirical evidence connecting social interaction behavior with financial outcomes. This work reports results from one of the largest known studies connecting financial outcomes and phone-based social behavior (180,000 individuals; 2 years' time frame; 82.2 million monthly bills, and 350 million call logs). Our methodology tackles highly imbalanced dataset, which is a pertinent problem with modelling credit risk behavior, and offers a novel hybrid method that yields improvements over, both, a traditional transaction data only approach, and an approach that uses only call data. The results pave way for better financial modelling of billions of unbanked and underbanked customers using non-traditional metrics like phone-based credit scoring.

What Do Stroke Patients Look for in Game-Based Rehabilitation: A Survey Study.

Stroke is one of the most common causes of physical disability, and early, intensive, and repetitive rehabilitation exercises are crucial to the recovery of stroke survivors. Unfortunately, research shows that only one third of stroke patients actually perform recommended exercises at home, because of the repetitive and mundane nature of conventional rehabilitation exercises. Thus, to motivate stroke survivors to engage in monotonous rehabilitation is a significant issue in the therapy process. Game-based rehabilitation systems have the potential to encourage patients continuing rehabilitation exercises at home. However, these systems are still rarely adopted at patients' places. Discovering and eliminating the obstacles in promoting game-based rehabilitation at home is therefore essential. For this purpose, we conducted a study to collect and analyze the opinions and expectations of stroke patients and clinical therapists. The study is composed of 2 parts: Rehab-preference survey - interviews to both patients and therapists to understand the current practices, challenges, and expectations on game-based rehabilitation systems; and Rehab-compatibility survey - a gaming experiment with therapists to elaborate what commercial games are compatible with rehabilitation. The study is conducted with 30 outpatients with stroke and 19 occupational therapists from 2 rehabilitation centers in Taiwan. Our surveys show that game-based rehabilitation systems can turn the rehabilitation exercises more appealing and provide personalized motivation for various stroke patients. Patients prefer to perform rehabilitation exercises with more diverse and fun games, and need cost-effective rehabilitation systems, which are often built on commodity hardware. Our study also sheds light on incorporating the existing design-for-fun games into rehabilitation system. We envision the results are helpful in developing a platform which enables rehab-compatible (i.e., existing, appropriately selected) games to be operated on commodity hardware and brings cost-effective rehabilitation systems to more and more patients' home for long-term recovery.

A social diffusion model with an application on election simulation.

Issues about opinion diffusion have been studied for decades. It has so far no empirical approach to model the interflow and formation of crowd's opinion in elections due to two reasons. First, unlike the spread of information or flu, individuals have their intrinsic attitudes to election candidates in advance. Second, opinions are generally simply assumed as single values in most diffusion models. However, in this case, an opinion should represent preference toward multiple candidates. Previously done models thus may not intuitively interpret such scenario. This work is to design a diffusion model which is capable of managing the aforementioned scenario. To demonstrate the usefulness of our model, we simulate the diffusion on the network built based on a publicly available bibliography dataset. We compare the proposed model with other well-known models such as independent cascade. It turns out that our model consistently outperforms other models. We additionally investigate electoral issues with our model simulator.