Gradient-Based Multi-Objective Feature Selection for Gait Mode Recognition of Transfemoral Amputees.

One control challenge in prosthetic legs is seamless transition from one gait mode to another. User intent recognition (UIR) is a high-level controller that tells a low-level controller to switch to the identified activity mode, depending on the user's intent and environment. We propose a new framework to design an optimal UIR system with simultaneous maximum performance and minimum complexity for gait mode recognition. We use multi-objective optimization (MOO) to find an optimal feature subset that creates a trade-off between these two conflicting objectives. The main contribution of this paper is two-fold: (1) a new gradient-based multi-objective feature selection (GMOFS) method for optimal UIR design; and (2) the application of advanced evolutionary MOO methods for UIR. GMOFS is an embedded method that simultaneously performs feature selection and classification by incorporating an elastic net in multilayer perceptron neural network training. Experimental data are collected from six subjects, including three able-bodied subjects and three transfemoral amputees. We implement GMOFS and four variants of multi-objective biogeography-based optimization (MOBBO) for optimal feature subset selection, and we compare their performances using normalized hypervolume and relative coverage. GMOFS demonstrates competitive performance compared to the four MOBBO methods. We achieve a mean classification accuracy of 97.14 % ± 1.51 % and 98.45 % ± 1.22 % with the optimal selected subset for able-bodied and amputee subjects, respectively, while using only 23% of the available features. Results thus indicate the potential of advanced optimization methods to simultaneously achieve accurate, reliable, and compact UIR for locomotion mode detection of lower-limb amputees with prostheses.

Neonatal Presentation of Unremitting Inflammatory Bowel Disease.

Very-early-onset inflammatory bowel disease (VEO-IBD) has a distinct phenotype and should be considered a specific entity. VEO-IBD presents with very severe clinical pictures and is frequently known by an indeterminate colitis whose clinical remission is unmanageable. This study examines the case of a neonate with VEO-IBD, not responding to medical and surgical treatment. A 7-day-old Iranian female neonate presented with severe bloody diarrhea, poor feeding, abdominal distention, and dehydration suggesting severe proctocolitis due to an allergy to the protein in cow's milk. The condition did not respond to the elimination of diet for 1 month. Infections, celiac disease, and cystic fibrosis were excluded. Immunological investigations were negative, but antineutrophil cytoplasmic antibodies were positive. Due to the neonate's persistent symptoms and failure to thrive, upper and lower endoscopies were performed, showing ulcerative colitis. At the age of 4 months, she presented with signs and symptoms of toxic colitis and acute intestinal perforation, which prompted an emergency laparotomy. Due to the necrosis of the colon, hemicolectomy and colostomy were done. The patient was resuscitated and rehabilitated and was given glucocorticoid and mesalamine. We believe that the incidence of this problem is increasing, as is shown by the rise in the number of children under 10 years old being diagnosed. These patients require more aggressive therapeutic interventions than older IBD patients to achieve complete remission because they are more likely to have extensive colonic disease.

A Successful implementation of an idea to a nationally approved plan: Analyzing Iran's National Health Roadmap using the Kingdon model of policymaking.

Introduction: Hospital beds, human resources, and medical equipment are the costliest elements in the health system and play an essential role at the time of treatment. In this paper, different phases of the NEDA 2026 project and its methodological approach were presented and its formulation process was analysed using the Kingdon model of policymaking. Methods: Iran Health Roadmap (NEDA 2026) project started in March 2016 and ended in March 2017. The main components of this project were hospital beds, clinical human resources, specialist personnel, capital medical equipment, laboratory facilities, emergency services, and service delivery model. Kingdon model of policymaking was used to evaluate NEDA 2026 development and implementation. In this study, all activities to accomplish each step in the Kingdon model was described. Results: The followings were done to accomplish the goals of each step: collecting experts' viewpoint (problem identification and definition), systematic review of the literature, analysis of previous experiences, stakeholder analysis, economic analysis, and feasibility study (solution appropriateness analysis), three-round Delphi survey (policy survey and scrutinization), and intersectoral and interasectoral agreement (policy legislation). Conclusion: In the provision of an efficient health service, various components affect each other and the desired outcome, so they need to be considered as parts of an integrated system in developing a roadmap for the health system. Thus, this study demonstrated the cooperation process at different levels of Iran's health system to formulate a roadmap to provide the necessary resources for the health sector for the next 10 years and to ensure its feasibility using the Kingdon policy framework.

Host metabolome predicts the severity and onset of acute toxicities induced by CAR T-cell therapy.

Anti-CD19 chimeric antigen receptor (CAR) T-cell therapy is a highly effective treatment option for patients with relapsed/refractory large B-cell lymphoma. However, widespread use is deterred by the development of clinically significant acute inflammatory toxicities, including cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS), that induce significant morbidity and require close monitoring. Identification of host biochemical signatures that predict the severity and time-to-onset of CRS and ICANS may assist patient stratification to enable timely mitigation strategies. Here, we report pretreatment host metabolites that are associated with CRS and ICANS induced by axicabtagene ciloleucel or tisagenlecleucel therapy. Both untargeted metabolomics analysis and validation using targeted assays revealed a significant association between the abundance of specific pretreatment biochemical entities and an increased risk and/or onset of clinically significant CRS (q < .1) and ICANS (q < .25). Higher pretreatment levels of plasma glucose and lower levels of cholesterol and glutamate were associated with a faster onset of CRS. In contrast, low baseline levels of the amino acids proline and glycine and the secondary bile acid isoursodeoxycholate were significantly correlated with clinically significant CRS. Lower concentration of the amino acid hydroxyproline was associated with higher grade and faster onset of ICANS, whereas low glutamine was negatively correlated with faster development of ICANS. Overall, our data indicate that the pretreatment host metabolome has biomarker potential in determining the risk of clinically significant CRS and ICANS, and may be useful in risk stratification of patients before anti-CD19 CAR T-cell therapy.

Identification of robust deep neural network models of longitudinal clinical measurements.

Deep learning (DL) from electronic health records holds promise for disease prediction, but systematic methods for learning from simulated longitudinal clinical measurements have yet to be reported. We compared nine DL frameworks using simulated body mass index (BMI), glucose, and systolic blood pressure trajectories, independently isolated shape and magnitude changes, and evaluated model performance across various parameters (e.g., irregularity, missingness). Overall, discrimination based on variation in shape was more challenging than magnitude. Time-series forest-convolutional neural networks (TSF-CNN) and Gramian angular field(GAF)-CNN outperformed other approaches (P < 0.05) with overall area-under-the-curve (AUCs) of 0.93 for both models, and 0.92 and 0.89 for variation in magnitude and shape with up to 50% missing data. Furthermore, in a real-world assessment, the TSF-CNN model predicted T2D with AUCs reaching 0.72 using only BMI trajectories. In conclusion, we performed an extensive evaluation of DL approaches and identified robust modeling frameworks for disease prediction based on longitudinal clinical measurements.

Toward Minimal-Sensing Locomotion Mode Recognition for a Powered Knee-Ankle Prosthesis.

OBJECTIVE: Locomotion mode recognition (LMR) enables seamless and natural transitions between low-level control systems in a powered prosthesis. We present a new optimization framework for LMR that eliminates irrelevant or redundant features and measurement signals while still maintaining performance. METHODS: We use multi-objective biogeography-based optimization to find a compromise solution between performance and the minimization of feature set size. Experimental data are collected from four transfemoral users walking with a powered knee-ankle prosthesis. We compare the performance of LMR systems trained with the optimal feature subsets and with the full feature set using a deep neural network classifier across six locomotion modes: standing, flat-ground walking, stair up/down, and ramp up/down. RESULTS: Statistical tests indicate that classifier performance using the optimal feature subsets is statistically equal to that using the full feature set. The LMR trained with an optimal subset results in the 1.98% steady-state and 4.09% transitional error rates, while only using approximately 41% and 53% of the available features and sensors, respectively. CONCLUSION: Results thus indicate the capability of the proposed framework to achieve simultaneously accurate and low-complex LMR systems for transfemoral individuals with powered prostheses. SIGNIFICANCE: This framework would potentially lead to less frequent clinical visits needed for sensor replacement and calibrations, which may save health care costs and the prosthesis user's time and energy.

The effects of implementation of a stepwise algorithmic protocol for nutrition care process in gastro-intestinal surgical children in Pediatric Intensive Care Unit (PICU).

INTRODUCTION: Malnutrition is known as one of the major health problems among critically ill children; optimum nutrition support is considered as a therapeutic strategy to improve clinical outcomes and minimize the length of Pediatric Intensive Care Unit (PICU) staying as well as its costs. Implementation of an algorithmic protocol can result in the upgrade of the quality of nutrition support system in PICU. METHOD: In this study, we developed a stepwise algorithmic nutrition care protocol for PICU patients in two phases as follows: a critical review of past literature and an expert discussion panel. The final structured protocol includes three following steps for the nutrition care process: 1) Initial nutrition screening and assessment, 2) Nutritional intervention and 3) Nutritional monitoring. Pre and post-implementation audits were carried out in a 23 bed medical/surgical PICU in a children's hospital affiliated to Mashhad University of Medical Sciences over two 4-week periods to evaluate the impact of the algorithm implementation. The post-implementation audit was performed by passing 12 weeks from the protocol implementation. Critically ill children aged between 1month and 10 years, and PICU length of stay>24 h who were in post gastrointestinal surgery state, were enrolled. RESULTS: Totally, 34 eligible critically ill gastrointestinal surgical children in post-implementation audit were compared with 30 patients with similar eligibility criteria in the pre-implementation audit. Notably, there were no significant differences in gender, median age, length of PICU stay, and mechanical ventilation requirement in the two audits. The comparison of our pre and post-implementation audits showed a significant increase in the proportion of energy delivery goal achievement following performing our intervention (56.7%, and 85.3%, for pre and post-implementation audits, respectively; p-value = 0.01).Additionally, time of achieving energy and protein goals were significantly decreased (5.5 vs. 3 days; p-value = 0.008 and 4 vs 3 days; p-value = 0.002, for energy and protein delivery goal achievements, respectively). CONCLUSION: The implementation of the algorithm have significantly improved the adequacy ratio of energy delivery and also decreased the time to achieve the goal in energy and protein intake among critically ill children in post-gastrointestinal state. In this regard, further prospective studies are needed for continuing the evaluation of the algorithm implementation outcomes in critically ill children with different surgical and internal underlying diseases.

Impact of Nutrition Support Team on Postoperative Nutritional Status and Outcome of Patients with Congenital Gastrointestinal Anomalies.

BACKGROUND The aim of this study was to evaluate postoperative nutritional status in patients who underwent operations due to congenital gastrointestinal anomalies in surgical neonatal intensive care units (NICUs) and to investigate the role of nutrition support teams (NSTs) on the outcome. METHODS A retrospective clinical study was carried out at two NICUs in Dr. Sheikh Pediatric Hospital, Mashhad, Iran. One of the NICUs was supported by NST and the other was not. A total of 120 patients were included through a non-random simple sampling. Different variables such as age, sex, prematurity, type of anomaly, birth weight, use of vasoactive drugs, weight gain in NICU, length of NICU stay, postoperative enteral nutrition initiation, duration of mechanical ventilation, mortality rate, maximum of blood sugar, the amount of calorie delivered to the calorie requirement ratio, and distribution of energy from enteral or parenteral roots were compared between the patients of two NICUs. RESULTS Median weight gain and the amount of calorie delivered during NICU stay in subjects of NSTsupported NICU was significantly more than other NICU. There was no significant difference in the length of NICU stay, enteral nutrition initiation after the operation, ventilation days, and percent of mortality between the two groups. The percentage of enteral feeding was also increased by about 2.8%, which was not significant. CONCLUSION NST could increase post-operative weight gain and calorie delivery in patients as well as providing an increase in enteral feeding rather than parenteral.

Optimal Mixed Tracking/Impedance Control With Application to Transfemoral Prostheses With Energy Regeneration.

OBJECTIVE: We design an optimal passivity-based tracking/impedance control system for a robotic manipulator with energy regenerative electronics, where the manipulator has both actively and semi-actively controlled joints. The semi-active joints are driven by a regenerative actuator that includes an energy-storing element. METHOD: External forces can have a large influence on energy regeneration characteristics. Impedance control is used to impose a desired relationship between external forces and deviation from reference trajectories. Multi-objective optimization (MOO) is used to obtain optimal impedance parameters and control gains to compromise between the two conflicting objectives of trajectory tracking and energy regeneration. We solve the MOO problem under two different scenarios: 1) constant impedance; and 2) time-varying impedance. RESULTS: The methods are applied to a transfemoral prosthesis simulation with a semi-active knee joint. Normalized hypervolume and relative coverage are used to compare Pareto fronts, and these two metrics show that time-varying impedance provides better performance than constant impedance. The solution with time-varying impedance with minimum tracking error (0.0008 rad) fails to regenerate energy (loses 9.53 J), while a solution with degradation in tracking (0.0452 rad) regenerates energy (gains 270.3 J). A tradeoff solution results in fair tracking (0.0178 rad) and fair energy regeneration (131.2 J). CONCLUSION: Our experimental results support the possibility of net energy regeneration at the semi-active knee joint with human-like tracking performance. SIGNIFICANCE: The results indicate that advanced control and optimization of ultracapacitor-based systems can significantly reduce power requirements in transfemoral prostheses.

Association of dysglycemia with mortality in children receiving parenteral nutrition in pediatric intensive care unit.

Khajavi L, Khademi G, Mehramiz M, Norouzy A, Safarian M. Association of dysglycemia with mortality in children receiving parenteral nutrition in pediatric intensive care unit. Turk J Pediatr 2018; 60: 134-141. One of the most important complications of parenteral nutrition (PN) is a high incidence of hyperglycemia. The aim of this study was to assess the effect of parenteral nutrition dysglycemia on clinical outcomes among critically ill children in pediatric intensive care unit (PICU). Charts of 201 critically ill children admitted in PICU during 2012-2015 were reviewed retrospectively. We included patients who were < 6 years of age and had received at least 60% of total energy from PN for a minimum of 5 days in PICU. The exclusion criteria were patients with diagnosis of diabetes mellitus, primary hypoglycemia, inborn errors of metabolism and patients who received dialyses. We defined hyperglycemia as blood glucose ≥150 mg/dl, and hypoglycemia as blood glucose ≤60 mg/dl. Based on blood glucose, patients were divided into four groups: `only hyperglycemia group` (having at least one hyperglycemia episode), `only hypoglycemia group` (having at least one hypoglycemia episode), `glucose variability` (having both hypoglycemia and hyperglycemia episodes), and `normoglycemia` (all glucose measurements were in normal range). Hyperglycemia and hypoglycemia occurred in 52.8% and 24.9% of all children, respectively; glucose variability occurred in 13.9% of all children. Multiple logistic regression analysis showed that glucose variability (OR: 3.1; 95% CI: 1.13-8.43) and hyperglycemia (OR: 2.14; 95% CI: 1.1-4.57) were associated with mortality independently. In `only hypoglycemia` group (N=22) there were only three deaths. There were no significant differences in the quantities of macronutrients prescribed via parenteral nutrition among the four blood glucose groups. Results of this study showed that hyperglycemia and glucose variability are strong predictors of mortality in pediatrics receiving parenteral nutrition.

Clinical outcome and bowel function after surgical treatment in Hirschsprung's disease.

BACKGROUND: Bowel function has been reported to be adversely affected following surgery in cases of Hirschsprung. We retrospectively studied both the clinical outcome and bowel function status following surgery in patients diagnosed with Hirschprung's disease (HD). 161 cases, who underwent pull-through operations for HD in Sheikh Pediatric Tertiary Centre, Mashhad, Iran. The specified time bracket spanned between 2006 and 2011. MATERIALS AND METHODS: Data was extracted from Health Information System with the aim of investigating patients for both short and long-term gastrointestinal (GI) complications after surgery bases in addition to the concurrence of any associated anomalies. Three main procedures were analysed in this respect (Swenson, Duhamel and Soave). RESULTS: In a study of 96 (59%) boys and 65 (40.3%) girls, mortality rate was reported to be 15.5% (15 males and 10 females). A considerable majority of almost three fourths were detected with both early and late GI complications after surgery. The latter mainly included constipation (30.8%), incontinence (19.8%), enterocolitis (8%), diarrhea (11%) in a declining order of incidence. Down syndrome and others HD-associated anomalies were detected in 3.7% and 24.3% of cases respectively. CONCLUSIONS: Constipation and foecal incontinence were the most prevalent postoperative complications, which were reported almost as frequent in other studies. Yet, Enterocolitis, was reported slightly less in prevalence. Also mortality rates were considerably higher, compared to developed nations.

Noise pollution in intensive care units and emergency wards.

INTRODUCTION: The improvement of technology has increased noise levels in hospital Wards to higher than international standard levels (35-45 dB). Higher noise levels than the maximum level result in patient's instability and dissatisfaction. Moreover, it will have serious negative effects on the staff's health and the quality of their services. The purpose of this survey is to analyze the level of noise in intensive care units and emergency wards of the Imam Reza Teaching Hospital, Mashhad. PROCEDURE: This research was carried out in November 2009 during morning shifts between 7:30 to 12:00. Noise levels were measured 10 times at 30-minute intervals in the nursing stations of 10 wards of the emergency, the intensive care units, and the Nephrology and Kidney Transplant Departments of Imam Reza University Hospital, Mashhad. The noise level in the nursing stations was tested for both the maximum level (Lmax) and the equalizing level (Leq). The research was based on the comparison of equalizing levels (Leq) because maximum levels were unstable. RESULTS: In our survey the average level (Leq) in all wards was much higher than the standard level. The maximum level (Lmax) in most wards was 85-86 dB and just in one measurement in the Internal ICU reached 94 dB. The average level of Leq in all wards was 60.2 dB. In emergency units, it was 62.2 dB, but it was not time related. The highest average level (Leq) was measured at 11:30 AM and the peak was measured in the Nephrology nursing station. CONCLUSION: The average levels of noise in intensive care units and also emergency wards were more than the standard levels and as it is known these wards have vital roles in treatment procedures, so more attention is needed in this area.