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.

Apply influence diagrams for utility analysis of paying the weight-reducing expenses: a case study in taiwan.

To effectively control the growth of medical expenditure, Bureau of National Health Insurance (NHI) of Taiwan has taken many measures, including the Reasonable Number of Outpatient Services, Ceiling Price, Global Budgets, Strategic Analysis and the Excellence Plan; however, these measures can only scratch the surface. Due to the change of life style and the deteriorating condition of over-nutrition and obesity, people now have a higher risk of diabetes, hypertension, hyperlipidemia, cardiovascular disease, gallbladder disease, cancer, gout, arthritis, and so on, which leads to higher medical expenditure. Therefore, good civil preventive health care is regarded as the solution of surging medical expenditure. According to NHI's statistics, the annual medical expenditure of diabetes is about 13 billion NT dollars. Among these diabetics, over 95% are affected by type 2 diabetes mellitus; at least two-thirds--over 80% according to some researches--are overweight or obese. The research says, losing 5% to 10% of the original body weight can lower the risk of chronic diseases effectively; also, giving early therapy for obesity can reduce the complication probability, thus for avoiding the waste of medical resources. By applying influence diagrams of Bayesian Network and Utility Expect of statistics, this paper evaluates the medical expenditure of Taiwan's NHI under the circumstances of providing and not providing benefit for weight-loss outpatient services. The result of this research is that the cost of not providing benefit for weight-loss outpatient services is 3.4 times of the contrary. Therefore, if Taiwan's NHI provides reasonable benefit for weight-loss outpatient services, not only the risk of people suffering from diabetes, hypertension, hyperlipidemia, cardiovascular disease, gallbladder disease, cancer, gout, arthritis, etc. will go down; but also the medical expenditure can be effectively reduced.

A study of pre-decision evaluation using influence diagram: an estimation of the benefits of influenza vaccination.

The purpose of this study is to develop a decision analysis model based on the influence diagram and estimate the benefits receiving of influenza vaccination. We collected more than 300,000 samples of elders aged over 65 years in Taiwan and then analyzed the health expenditure of the elders with and without influenza vaccination. We incorporate clinical results and the knowledge of physicians by an influence diagram. We divided our samples into four different age groups and the results showed that the total healthcare expenses for receiving influenza vaccination are more than the expenses for not receiving influenza vaccination for all age groups, we found there is a trend that the difference decreases if the age is older. We performed the one-way sensitivity analysis and Monde Carlo sensitivity analysis further and the results showed that the expected health expenditure is mostly sensitive to the hospitalization under the different condition.

A new method to guard inpatient medication safety by the implementation of RFID.

Since life is invaluable, the patient safety is always an important issue. How to reduce the malpractices and advance the patient safety is the primary goal of many countries. The current problem is that the hospitals cannot quickly and precisely identify the name of medicine, the position of patient and staff and the servicing time and dosage taken by patients. The application of Radio Frequency Identification (RFID) is rocketing in popularity as varieties of expanded uses. However, due to the investment consideration, there are few cases that practically implement such a technology in healthcare industries. This paper presents a Wisely Aware RFID Dosage (WARD) system, which based on an integration of barcodes and RFID tags, to demonstrate effective and safe patient care environment, for preventing the risk of medication error. Finally, through an evaluation of users' satisfaction, a reliability of 0.92 and a criterion-related validity of 0.82 show that this system is able to effectively construct the patient-safety-centric environment.