Unlocking stroke prediction: Harnessing projection-based statistical feature extraction with ML algorithms.

Non-communicable diseases, such as cardiovascular disease, cancer, chronic respiratory diseases, and diabetes, are responsible for approximately 71% of all deaths worldwide. Stroke, a cerebrovascular disorder, is one of the leading contributors to this burden among the top three causes of death. Early recognition of symptoms can encourage a balanced lifestyle and provide essential information for stroke prediction. To identify a stroke patient and risk factors, machine learning (ML) is a key tool for physicians. Due to different data measurement scales and their probability distributional assumptions, ML-based algorithms struggle to detect risk factors. Furthermore, when dealing with risk factors with high-dimensional features, learning algorithms struggle with complexity. In this study, rigorous statistical tests are used to identify risk factors, and PCA-FA (Integration of Principal Components and Factors) and FPCA (Factor Based PCA) approaches are proposed for projecting suitable feature representations for improving learning algorithm performances. The study dataset consists of different clinical, lifestyle, and genetic attributes, allowing for a comprehensive analysis of potential risk factors associated with stroke, which contains 5110 patient records. Using significant test (P-value <0.05), chi-square and independent sample t-test identified age, heart_disease, hypertension, work_type, ever_married, bmi, and smoking_status as risk factors for stroke. To develop the predicting model with proposed feature extraction techniques, random forests approach provides the best results when utilizing the PCA-FA method. The best accuracy rate for this approach is 92.55%, while the AUC score is 98.15%. The prediction accuracy has increased from 2.19% to 19.03% compared to the existing work. Additionally, the prediction results is robustified and reproducible with a stacking ensemble-based classification algorithm. We also developed a web-based application to help doctors diagnose stroke risk based on the findings of this study, which could be used as an additional tool to help doctors diagnose.

Improving diabetes disease patients classification using stacking ensemble method with PIMA and local healthcare data.

Diabetes mellitus, a chronic metabolic disorder, continues to be a major public health issue around the world. It is estimated that one in every two diabetics is undiagnosed. Early diagnosis and management of diabetes can also prevent or delay the onset of complications. With the help of a variety of machine learning and deep learning models, stacking algorithms, and other techniques, our study's goal is to detect diseases early. In this study, we propose two stacking-based models for diabetes disease classification using a combination of the PIMA Indian diabetes dataset, simulated data, and additional data collected from a local healthcare facility. We use both the classical and deep neural network stacking ensemble methods to combine the predictions of multiple classification models and improve classification accuracy and robustness. In the evaluation protocol, we used both the train-test and cross-validation (CV) techniques to validate our proposed model. The highest accuracy is obtained by stacking ensemble with three NN architectures, resulting in an accuracy of 95.50 %, precision of 94 %, recall of 97 %, and f1-score of 96 % using 5-fold CV on simulation study. The stacked accuracy obtained from ML algorithms for the Pima Indian Diabetes dataset is 75.03 % using the train-test split protocol, while the accuracy obtained from the CV protocol is 77.10 % on the stacked model. The range of performance scores that outperformed the CV protocol 2.23 %-12 %. Our proposed method achieves a high accuracy range from 92 % to 95 %, precision, recall, and F1-score ranges from 88 % to 96 % using classical and deep neural network (NN)-based stacking method on the primary dataset. The proposed dataset and ensemble method could be useful in the early detection and treatment of diabetes, as well as in the advancement of machine learning and data analysis techniques in the healthcare industry.

Boosting efficiency above 28% using effective charge transport layer with Sr3SbI3 based novel inorganic perovskite.

Strontium antimony iodide (Sr3SbI3) is one of the emerging absorbers materials owing to its intriguing structural, electronic, and optical properties for efficient and cost-effective solar cell applications. A comprehensive investigation on the structural, optical, and electronic characterization of Sr3SbI3 and its subsequent applications in heterostructure solar cells have been studied theoretically. Initially, the optoelectronic parameters of the novel Sr3SbI3 absorber, and the possible electron transport layer (ETL) of tin sulfide (SnS2), zinc sulfide (ZnS), and indium sulfide (In2S3) including various interface layers were obtained by DFT study. Afterward, the photovoltaic (PV) performance of Sr3SbI3 absorber-based cell structures with SnS2, ZnS, and In2S3 as ETLs were systematically investigated at varying layer thickness, defect density bulk, doping density, interface density of active materials including working temperature, and thereby, optimized PV parameters were achieved using SCAPS-1D simulator. Additionally, the quantum efficiency (QE), current density-voltage (J-V), and generation and recombination rates of photocarriers were determined. The maximum power conversion efficiency (PCE) of 28.05% with JSC of 34.67 mA cm-2, FF of 87.31%, VOC of 0.93 V for SnS2 ETL was obtained with Al/FTO/SnS2/Sr3SbI3/Ni structure, while the PCE of 24.33%, and 18.40% in ZnS and In2S3 ETLs heterostructures, respectively. The findings of this study contribute to in-depth understanding of the physical, electronic, and optical properties of Sr3SbI3 absorber perovskite and SnS2, ZnS, and In2S3 ETLs. Additionally, it provides valuable insights into the potential of Sr3SbI3 in heterostructure perovskite solar cells (PSCs), paving the pathway for further experimental design of an efficient and stable PSC devices.

Engineered Amber-Emitting Nano Luciferase and Its Use for Immunobioluminescence Imaging In Vivo.

The NanoLuc luciferase (NLuc) and its furimazine (FRZ) substrate have revolutionized bioluminescence (BL) assays and imaging. However, the use of the NLuc-FRZ luciferase-luciferin pair for mammalian tissue imaging is hindered by the low tissue penetration of the emitting blue photons. Here, we present the development of an NLuc mutant, QLuc, which catalyzes the oxidation of a synthetic QTZ luciferin for bright and red-shifted emission peaking at ∼585 nm. Compared to other small single-domain NLuc mutants, this amber-light-emitting luciferase exhibited improved performance for imaging deep-tissue targets in live mice. Leveraging this novel bioluminescent reporter, we further pursued in vivo immunobioluminescence imaging (immunoBLI), which used a fusion protein of a single-chain variable antibody fragment (scFv) and QLuc for molecular imaging of tumor-associated antigens in a xenograft mouse model. As one of the most red-shifted NLuc variants, we expect QLuc to find broad applications in noninvasive mammalian imaging. Moreover, the immunoBLI method complements immunofluorescence imaging and immuno-positron emission tomography (immunoPET), serving as a convenient and nonradioactive molecular imaging tool for animal models in basic and preclinical research.

Fluorophore-NanoLuc BRET Reporters Enable Sensitive In Vivo Optical Imaging and Flow Cytometry for Monitoring Tumorigenesis.

Fluorescent proteins are widely used to study molecular and cellular events, yet this traditionally relies on delivery of excitation light, which can trigger autofluorescence, photoxicity, and photobleaching, impairing their use in vivo. Accordingly, chemiluminescent light sources such as those generated by luciferases have emerged, as they do not require excitation light. However, current luciferase reporters lack the brightness needed to visualize events in deep tissues. We report the creation of chimeric eGFP-NanoLuc (GpNLuc) and LSSmOrange-NanoLuc (OgNLuc) fusion reporter proteins coined LumiFluors, which combine the benefits of eGFP or LSSmOrange fluorescent proteins with the bright, glow-type bioluminescent light generated by an enhanced small luciferase subunit (NanoLuc) of the deep-sea shrimp Oplophorus gracilirostris. The intramolecular bioluminescence resonance energy transfer that occurs between NanoLuc and the fused fluorophore generates the brightest bioluminescent signal known to date, including improved intensity, sensitivity, and durable spectral properties, thereby dramatically reducing image acquisition times and permitting highly sensitive in vivo imaging. Notably, the self-illuminating and bifunctional nature of these LumiFluor reporters enables greatly improved spatiotemporal monitoring of very small numbers of tumor cells via in vivo optical imaging and also allows the isolation and analyses of single cells by flow cytometry. Thus, LumiFluor reporters are inexpensive, robust, noninvasive tools that allow for markedly improved in vivo optical imaging of tumorigenic processes.

Ribosomal synthesis of backbone macrocyclic peptides.

A wealth of knowledge has been accumulated on ribosomal synthesis of macrocyclic peptides in the past decade. In nature, backbone cyclization of the translated linear peptides is generally catalyzed by specific enzymes, giving them peptidase resistance, thermodynamic stability and various other physiological activities. Due to these biochemical traits, backbone cyclic peptides have become an attractive resource for the discovery of drug leads. Recently, various new methodologies have also been established to generate man-made cyclic peptides. Here, we describe the biosynthetic mechanisms of naturally occurring backbone macrocyclic peptides focusing on cyclotides, sunflower trypsin inhibitors (SFTIs) and cyanobactins as well as several new emerging methodologies, such as sortase mediated ligation, protein splicing method and genetic code reprogramming.