Enhanced Runge-Kutta-driven feature selection model for early detection of gastroesophageal reflux disease.

Gastroesophageal reflux disease (GERD) profoundly compromises the quality of life, with prolonged untreated cases posing a heightened risk of severe complications such as esophageal injury and esophageal carcinoma. The imperative for early diagnosis is paramount in averting progressive pathological developments. This study introduces a wrapper-based feature selection model based on the enhanced Runge Kutta algorithm (SCCRUN) and fuzzy k-nearest neighbors (FKNN) for GERD prediction, named bSCCRUN-FKNN-FS. Runge Kutta algorithm (RUN) is a metaheuristic algorithm designed based on the Runge-Kutta method. However, RUN's effectiveness in local search capabilities is insufficient, and it exhibits insufficient convergence accuracy. To enhance the convergence accuracy of RUN, spiraling communication and collaboration (SCC) is introduced. By facilitating information exchange among population individuals, SCC expands the solution search space, thereby improving convergence accuracy. The optimization capabilities of SCCRUN are experimentally validated through comparisons with classical and state-of-the-art algorithms on the IEEE CEC 2017 benchmark. Subsequently, based on SCCRUN, the bSCCRUN-FKNN-FS model is proposed. During the period from 2019 to 2023, a dataset comprising 179 cases of GERD, including 110 GERD patients and 69 healthy individuals, was collected from Zhejiang Provincial People's Hospital. This dataset was utilized to compare our proposed model against similar algorithms in order to evaluate its performance. Concurrently, it was determined that features such as the internal diameter of the esophageal hiatus during distention, esophagogastric junction diameter during distention, and external diameter of the esophageal hiatus during non-distention play crucial roles in influencing GERD prediction. Experimental findings demonstrate the outstanding performance of the proposed model, with a predictive accuracy reaching as high as 93.824 %. These results underscore the significant advantage of the proposed model in both identifying and predicting GERD patients.

Wearable-Sensor-Based Weakly Supervised Parkinson's Disease Assessment with Data Augmentation.

Parkinson's disease (PD) is the second most prevalent dementia in the world. Wearable technology has been useful in the computer-aided diagnosis and long-term monitoring of PD in recent years. The fundamental issue remains how to assess the severity of PD using wearable devices in an efficient and accurate manner. However, in the real-world free-living environment, there are two difficult issues, poor annotation and class imbalance, both of which could potentially impede the automatic assessment of PD. To address these challenges, we propose a novel framework for assessing the severity of PD patient's in a free-living environment. Specifically, we use clustering methods to learn latent categories from the same activities, while latent Dirichlet allocation (LDA) topic models are utilized to capture latent features from multiple activities. Then, to mitigate the impact of data imbalance, we augment bag-level data while retaining key instance prototypes. To comprehensively demonstrate the efficacy of our proposed framework, we collected a dataset containing wearable-sensor signals from 83 individuals in real-life free-living conditions. The experimental results show that our framework achieves an astounding 73.48% accuracy in the fine-grained (normal, mild, moderate, severe) classification of PD severity based on hand movements. Overall, this study contributes to more accurate PD self-diagnosis in the wild, allowing doctors to provide remote drug intervention guidance.

Unilateral Biportal Endoscopic Posterior Cervical Foraminotomy: An Outcome Comparison With the Full-endoscopic Posterior Cervical Foraminotomy.

STUDY DESIGN: Retrospective control study. OBJECTIVE: To compare the curative effects of unilateral biportal endoscopic posterior cervical foraminotomy (UBE-PCF) with full-endoscopic posterior cervical foraminotomy (FPCF). SUMMARY OF BACKGROUND DATA: There are few studies directly comparing outcomes between UBE-PCF and FPCF. The objective of this study was to compare outcomes between UBE-PCF and FPCF. METHODS: A retrospective control study was conducted for 69 patients of cervical radiculopathy from July 2019 to December 2021. Clinical outcomes scores, including neck disability index, visual analog scale (VAS)-arm, and VAS-neck were evaluated. Serum creatine kinase levels and the size of the operating hole were measured. RESULTS: Postoperative neck disability index, VAS-neck, and VAS-arm scores showed statistically significant improvement over preoperative scores ( P <0.01). The operating time was significantly shorter in the UBE-PCF group ( P <0.001). No significant differences were found in serum creatine kinase levels between the 2 groups ( P >0.05). The mean area of the operating hole was 1.47+0.05 cm 2 in the FPCF group and 1.79+0.11 cm 2 in the UBE-PCF group. The difference was statistically significant ( P <0.001). CONCLUSIONS: Both UBE-PCF and FPCF are safe and effective procedures for cervical radiculopathy. Predictable and sufficient decompression could be achieved by UBE-PCF in a shorter operation time. LEVEL OF EVIDENCE: Treatment Benefits Level III.

Efficient multi-task learning with adaptive temporal structure for progression prediction.

In this paper, we propose a novel efficient multi-task learning formulation for the class of progression problems in which its state will continuously change over time. To use the shared knowledge information between multiple tasks to improve performance, existing multi-task learning methods mainly focus on feature selection or optimizing the task relation structure. The feature selection methods usually fail to explore the complex relationship between tasks and thus have limited performance. The methods centring on optimizing the relation structure of tasks are not capable of selecting meaningful features and have a bi-convex objective function which results in high computation complexity of the associated optimization algorithm. Unlike these multi-task learning methods, motivated by a simple and direct idea that the state of a system at the current time point should be related to all previous time points, we first propose a novel relation structure, termed adaptive global temporal relation structure (AGTS). Then we integrate the widely used sparse group Lasso, fused Lasso with AGTS to propose a novel convex multi-task learning formulation that not only performs feature selection but also adaptively captures the global temporal task relatedness. Since the existence of three non-smooth penalties, the objective function is challenging to solve. We first design an optimization algorithm based on the alternating direction method of multipliers (ADMM). Considering that the worst-case convergence rate of ADMM is only sub-linear, we then devise an efficient algorithm based on the accelerated gradient method which has the optimal convergence rate among first-order methods. We show the proximal operator of several non-smooth penalties can be solved efficiently due to the special structure of our formulation. Experimental results on four real-world datasets demonstrate that our approach not only outperforms multiple baseline MTL methods in terms of effectiveness but also has high efficiency.

A boost of thermoelectric generation performance for polycrystalline InTe by texture modulation.

The new rising binary InTe displays advantageously high electronic conductivity and low thermal conductivity along the [110] direction, providing a high potential of texture modulation for thermoelectric performance improvement. In this work, coarse crystalline InTe material with a high degree of texture along the [110] direction was realized by the oriented crystal hot-deformation method. The coarse grains with high texture not only maintain the preferred orientation of the zone-melting crystal as far as possible, but also greatly depress the grain boundary scattering, thus leading to the highest room temperature power factor of 8.7 µW cm-1 K-1 and a high average figure of merit of 0.71 in the range of 300-623 K. Furthermore, the polycrystalline characteristic with refined grains also promotes the mechanical properties. As a result, an 8-couple thermoelectric generator module consisting of p-type InTe and commercial n-type Bi2Te2.7Se0.3 legs was successfully integrated and a high conversion efficiency of ∼5.0% under the temperature difference of 290 K was achieved, which is comparable to traditional Bi2Te3 based modules. This work not only demonstrates the potential of InTe as a power generator near room temperature, but also provides one more typical example of a texture modulation strategy beyond the traditional Bi2Te3 thermoelectrics.

High Thermoelectric Performance of p-Type Bi0.4Sb1.6Te3+x Synthesized by Plasma-Assisted Ball Milling.

The exploration of new synthesis methods is important for the improvement of the thermoelectric property of a material for the different mechanisms of microstructure fabrication, surface activity modulation, and particle refinement. Herein, we prepared p-Bi2Te3 bulk materials by a simple synthesis method of the plasma-assisted ball milling, which yielded finer nanopowders, higher texture of in-plane direction, and higher efficiency compared to the traditional ball milling, favoring the simultaneous improvement of electrical and thermal properties. When combined with the Te liquid sintering, nano-/microscale hierarchical pores were fabricated and the carrier mobility was also increased, which together resulted in the low lattice thermal conductivity of 0.52 W·m-1·K-1 and the high power factor of 43.4 µW·cm-1·K-2 at 300 K, as well as the ranking ahead zT of 1.4@375 K. Thus, this work demonstrated the advantages of plasma-assisted ball milling in highly efficient synthesis of p-type Bi2Te3 with promising thermoelectric performance, which can also be utilized to prepare other thermoelectric materials.

A Multi-classification Accessment Framework for Reproducible Evaluation of Multimodal Learning in Alzheimer's Disease.

Multimodal learning is widely used in automated early diagnosis of Alzheimer's disease. However, the current studies are based on an assumption that different modalities can provide more complementary information to help classify the samples from the public dataset Alzheimer's Disease Neuroimaging Initiative (ADNI). In addition, the combination of modalities and different tasks are external factors that affect the performance of multimodal learning. Above all, we summrise three main problems in the early diagnosis of Alzheimer's disease: (i) unimodal vs multimodal; (ii) different combinations of modalities; (iii) classification of different tasks. In this paper, to experimentally verify these three problems, a novel and reproducible multi-classification framework for Alzheimer's disease early automatic diagnosis is proposed to evaluate and verify the above issues. The multi-classification framework contains four layers, two types of feature representation methods, and two types of models to verify these three issues. At the same time, our framework is extensible, that is, it is compatible with new modalities generated by new technologies. Following that, a series of experiments based on the ADNI-1 dataset are conducted and some possible explanations for the early diagnosis of Alzheimer's disease are obtained through multimodal learning. Experimental results show that SNP has the highest accuracy rate of 57.09% in the early diagnosis of Alzheimer's disease. In the modality combination, the addition of Single Nucleotide Polymorphism modality improves the multi-modal machine learning performance by 3% to 7%. Furthermore, we analyse and discuss the most related Region of Interest and Single Nucleotide Polymorphism features of different modalities.

Spatio-temporal Tensor Multi-Task Learning for Predicting Alzheimer's Disease in a Longitudinal study.

The utilisation of machine learning techniques to predict Alzheimer's Disease (AD) progression will substantially assist researchers and clinicians in establishing effective AD prevention and treatment strategies. In this research, we present a novel Multi-Task Learning (MTL) model for modelling AD progression based on tensor formation from spatio-temporal similarity measures of brain biomarkers. In this model, each patient sample's prediction in the tensor is assigned to a task, with each task sharing a set of latent factors acquired via tensor decomposition. To further improve the performance of the model, we present a novel regularisation term which utilises the convex combination of disease progression to modify longitudinal stability and ensure that two regression models have a minimal variation at successive time points. The model can be utilised to effectively predict AD progression with magnetic resonance imaging (MRI) data and cognitive scores of AD patients at various stages. We conducted extensive experiments to evaluate the performance for the proposed model and algorithm utilising data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Compared to single-task and state-of-the-art multi-task regression techniques, our proposed method has greater accuracy and stability for predicting AD progress in terms of root mean square error, with an average reduction of 2.60 compared to single-task regression methods and 1.17 compared to multi-task regression methods in the Mini-Mental State Examination (MMSE) questionnaire; with an average reduction of 5.08 compared to single-task regression methods and 2.71 compared to multi-task regression methods in the Alzheimer's Disease Assessment Scale-Cognitive subscale (ADAS-Cog).