Sentence Tokenization Using Statistical Unsupervised Machine Learning and Rule-Based Approach for Running Text in Gujarati Language

Background Several mapping systems are being introduced to guide atrial fibrillation (AF) ablation to patient-specific regions of interest. However, results have been extremely heterogeneous between studies, ranging from very poor to very promising. It is unknown if this reflects specific patient characteristics or procedural factors because most prior series were middle sized (N∼30–100 patients). Purpose To study 1 year and 3 year very long-term outcomes from map guided AF-driver ablation in a large patient registry with multiple operators, to identify clinical and procedural features influencing outcomes. In real-world AF patients with diverse comorbidities, we applied a consistent patient-tailored AF mapping and ablation strategy, monitored outcomes carefully and applied statistical and unsupervised machine learning approaches to identify features of success and failure. Method We studied 632 consecutive patients (65±10 y, 178 F) undergoing ablation for drug-refractory AF. 59.7% had persistent AF, and 29.9% had prior unsuccessful ablation (median 1 procedure). All patients underwent pulmonary vein isolation (PVI), followed by ablation of AF regions of interest mapped from 64 pole baskets (RhythmView, Abbott, IL), by 11 operators. Patients were followed using ambulatory ECG monitors quarterly for one year, and at the time of symptoms for 3 years. Results Fig. 1A shows overall freedom from AF at 1-year of 77.5% (95% CI: 74.2%, 80.9%) and at 3 years of 55.5% (95% CI: 51.2%, 60.1%). Freedom from AF/AT at 1-year was 70.1% (95% CI: 66.5%, 73.8%), and at 3 years was 48.6% (95% CI: 44.3%, 53.3%). Success was higher in patients with procedural termination, first ablation versus prior unsuccessful procedures, for paroxysmal AF than non-paroxysmal AF (1 year: AT/AF freedom 74.9% versus 66.7%, p=0.006), and smaller left atrium. Three clusters (Fig 1B) were identified comprising CHA2DS2VASc score, enlarged LA, prior failed case, presenting rhythm and termination during the procedure (Table 2). At 1 year, freedom from AT/AF was 77.8% (95% CI: 72.2%, 82.1%) for cluster 3 and 56.2% (95% CI: 48.3%, 65.4%) for cluster 1 (Fig. 1B). Conclusion In our large registry of N=632 patients undergoing AF-map guided ablations, machine learned clusters identified cohorts with success of 56.2 to 77.8% at 1 year. Future studies should identify if lower success represents technical challenges, such as difficulties in mapping very large atria, or more difficult to treat mechanisms. These results may inform patient inclusion and ablation strategy in upcoming AF treatment trials. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): National budget only - NIH, R01 HL149134, R01HL83359

Association between childhood trauma, parental bonding and antisocial personality disorder in adulthood: A machine learning approach

Inspired by natural language processing techniques, we here introduce Mol2vec, which is an unsupervised machine learning approach to learn vector representations of molecular substructures. Like the Word2vec models, where vectors of closely related words are in close proximity in the vector space, Mol2vec learns vector representations of molecular substructures that point in similar directions for chemically related substructures. Compounds can finally be encoded as vectors by summing the vectors of the individual substructures and, for instance, be fed into supervised machine learning approaches to predict compound properties. The underlying substructure vector embeddings are obtained by training an unsupervised machine learning approach on a so-called corpus of compounds that consists of all available chemical matter. The resulting Mol2vec model is pretrained once, yields dense vector representations, and overcomes drawbacks of common compound feature representations such as sparseness and bit collisions. The prediction capabilities are demonstrated on several compound property and bioactivity data sets and compared with results obtained for Morgan fingerprints as a reference compound representation. Mol2vec can be easily combined with ProtVec, which employs the same Word2vec concept on protein sequences, resulting in a proteochemometric approach that is alignment-independent and thus can also be easily used for proteins with low sequence similarities.

The aim of this study was to examine the risk factors for pharyngeal carriage of meningococci in third-level students using an unsupervised machine learning approach. Data were gathered as part of meningococcal prevalence studies conducted by the Irish Meningitis and Sepsis Reference Laboratory (IMSRL). Pharyngeal swab cultures for meningococcal carriage were taken from each student once they had completed a single-page anonymous questionnaire addressing basic demographics, social behaviors, living arrangements, vaccination, and antibiotic history. Data were analyzed using multiple correspondence analysis through a machine learning approach. In total, 16,285 students who had a pharyngeal throat swab taken returned a fully completed questionnaire. Overall, meningococcal carriage rate was 20.6%, and the carriage of MenW was 1.9% (n = 323). Young Irish adults aged under 20 years and immunized with the meningococcal C vaccine had a higher MenW colonization rate (n = 171/1260, 13.5%) compared with non-Irish adults aged 20 years or older without the MenC vaccine (n = 5/81, 6%, chi-square = 3.6, p = .05). Unsupervised machine learning provides a useful technique to explore meningococcal carriage risk factors. The issue is very complex, and asked risk factors only explain a small proportion of the carriage. This technique could be used on other conditions to explore reasons for carriage.

Human emotions can be expressed in many ways through facial expressions, speech, actions, and in textual form. Emotions play an important role in determining the human’s state and influence their lives in decision-making, behaviour and interaction. With the evolution of many advances in technology, researchers are able to detect the emotion of humans through facial expressions, speech, and text. Much advanced and efficient research work has been done in the field of emotion recognition through facial expression and audio, but still, there are not great works on emotion detection from text. Detecting emotions from text has wide applications in stock markets, business, decision-making, analyzing the view of people on any topic through social media conversations, tweets, blogs, and articles. Our work is focused on the detection of emotions from the text. We propose different approaches for detecting emotions through text. We implemented different approaches like the lexicon-based approach, the supervised machine learning approach with the Naïve-bayes algorithm, and the unsupervised machine learning approach with semantic similarities. After the consideration of the cons of all the approaches and the accuracies derived from our work, it is evident that the unsupervised machine learning approach with semantic similarities is the best approach with an accuracy of 78.5%.

A review on omics-based biomarkers discovery for Alzheimer's disease from the bioinformatics perspectives: Statistical approach vs machine learning approach

The contemporary scope of nuclear nephrology extends from non-imaging techniques for glomerular filtration rate calculation through dynamic renal scintigraphy and cortical imaging with planar or single photon emission computed tomography (SPECT) approaches to emerging applications in positron emission tomography (PET) renography and theranostics-based renal toxicity risk assessment. Artificial intelligence (AI) shares a long history with nuclear nephrology that started with expert systems and statistical machine learning (ML) approaches, transitioned through feed forward neural networks (FFNN), landed with convolutional neural networks (CNNs) and deep learning (DL), and has emerging opportunities across the gamut of generative AI like large language models (LLMs), diffusion models, generative adversarial networks (GANs) and multimodal models like vision language models (VLMs). A range of AI tools across the nuclear nephrology ecosystem describe bespoke AI algorithms, commercial AI products, embedded AI tools from vendors, general-purpose and cross-domain AI frameworks. Applications in clinical workflow, research and development, and imaging are explored, highlighting the potential of AI in detection, classification, segmentation, prediction, data analysis and image enhancement. Emerging AI opportunities from generative AI, LLMs, VLMs, and segmentation foundation models such as the SAM, offer exciting multi-modal, few-shot learning that may re-imagine nuclear nephrology. There remains the need for considerable development and validation for widespread clinical utility of AI opportunities, and the need for consideration of ethical limitations and social justice.

BackgroundThe rapid growth of the biomedical literature makes identifying strong evidence a time-consuming task. Applying machine learning to the process could be a viable solution that limits effort while maintaining accuracy.ObjectiveThe goal of the research was to summarize the nature and comparative performance of machine learning approaches that have been applied to retrieve high-quality evidence for clinical consideration from the biomedical literature.MethodsWe conducted a systematic review of studies that applied machine learning techniques to identify high-quality clinical articles in the biomedical literature. Multiple databases were searched to July 2020. Extracted data focused on the applied machine learning model, steps in the development of the models, and model performance.ResultsFrom 3918 retrieved studies, 10 met our inclusion criteria. All followed a supervised machine learning approach and applied, from a limited range of options, a high-quality standard for the training of their model. The results show that machine learning can achieve a sensitivity of 95% while maintaining a high precision of 86%.ConclusionsMachine learning approaches perform well in retrieving high-quality clinical studies. Performance may improve by applying more sophisticated approaches such as active learning and unsupervised machine learning approaches.

BackgroundThe main challenge in the management of indeterminate incidentally discovered adrenal tumours is to differentiate benign from malignant lesions. In the absence of clear signs of invasion or metastases, imaging techniques do not always precisely define the nature of the mass. The present pilot study aimed to determine whether radiomics may predict malignancy in adrenocortical tumours.MethodsCT images in unenhanced, arterial, and venous phases from 19 patients who had undergone resection of adrenocortical tumours and a cohort who had undergone surveillance for at least 5 years for incidentalomas were reviewed. A volume of interest was drawn for each lesion using dedicated software, and, for each phase, first-order (histogram) and second-order (grey-level colour matrix and run-length matrix) radiological features were extracted. Data were revised by an unsupervised machine learning approach using the K-means clustering technique.ResultsOf operated patients, nine had non-functional adenoma and 10 carcinoma. There were 11 patients in the surveillance group. Two first-order features in unenhanced CT and one in arterial CT, and 14 second-order parameters in unenhanced and venous CT and 10 second-order features in arterial CT, were able to differentiate adrenocortical carcinoma from adenoma (P < 0.050). After excluding two malignant outliers, the unsupervised machine learning approach correctly predicted malignancy in seven of eight adrenocortical carcinomas in all phases.ConclusionRadiomics with CT texture analysis was able to discriminate malignant from benign adrenocortical tumours, even by an unsupervised machine learning approach, in nearly all patients.

Stuck pipe during Geothermal well drilling is one of the problems that affect non-productive time (NPT) which prolong the drilling schedule. This problem directly increases drilling cost due to increasing overall drilling time, consumables, mud, drill bit and as well as sacrifices its well when there is no more progress to reach the targeted reservoir depth. Also the stuck pipe problem will create a domino effect that might impact next well drilling and another site drilling schedule. Stuck pipe problem is made from many contributor factors and also related to other problem cases such as lost circulation, wellbore instability, temporary zone closure, and cementing job trouble. The current analytical method mostly uses one or two dominant variables to predict stuck pipe problems which limitedly represents drilling conditions, lithology and wellbore conditions. A combination of statistical method and machine learning approach is developed and comprehensively used to predict stuck pipe problem that represents different well lithology, wellbore and drilling condition. Historical data that represents the conditional situation and electronic data recorder is used to predict the stuck pipe problem which different for each of the well. Many factors contribute to successful drilling activities and no one single factor, which dominantly becomes the root cause of the problem. A statistical approach such as principal component analysis, logistic regression and discriminant analysis is very helpful to interpret data before the next step of the analysis. Those statistical analyses should be the baseline of data interpretation before creating an algorithm to use predictive analysis using machine learning and deep learning.

Chaotic tumbleweed optimization algorithm with stacked deep learning based cyberattack detection in industrial CPS environment

Short-term load forecasting (STLF) is challenging due to complex time series (TS) which express three seasonal patterns and a nonlinear trend. This article proposes a novel hybrid hierarchical deep-learning (DL) model that deals with multiple seasonality and produces both point forecasts and predictive intervals (PIs). It combines exponential smoothing (ES) and a recurrent neural network (RNN). ES extracts dynamically the main components of each individual TS and enables on-the-fly deseasonalization, which is particularly useful when operating on a relatively small dataset. A multilayer RNN is equipped with a new type of dilated recurrent cell designed to efficiently model both short and long-term dependencies in TS. To improve the internal TS representation and thus the model's performance, RNN learns simultaneously both the ES parameters and the main mapping function transforming inputs into forecasts. We compare our approach against several baseline methods, including classical statistical methods and machine learning (ML) approaches, on STLF problems for 35 European countries. The empirical study clearly shows that the proposed model has high expressive power to solve nonlinear stochastic forecasting problems with TS including multiple seasonality and significant random fluctuations. In fact, it outperforms both statistical and state-of-the-art ML models in terms of accuracy.

In software industry estimation of effort, cost (EDC) and duration is a troublesome procedure.The exertion itself is in charge of experiencing trouble in evaluating EDC.In any software estimation process, the preeminent advance is to characterize and comprehends the framework to be assessed.Software cost estimation algorithmic methods are estimating by analogy, expert judgment method, price to win method, topdown method, bottom-up method are developed by researchers in the field of software engineering.Any methods are not superior to the other methods.In fact, the qualities and shortcomings of those strategies are differentiating to one another.Presently there are two types of software EDC models to estimate the software those are statistical approaches and machine learning approaches.Many of the software cost estimation methods follows statistical approaches, which are not having the capability to present causes and strong or accurate results.Machine learning techniques are suitable in software engineering because it produces accurate results, which estimates by training rules and iterations.Machine learning techniques resolve the challenges like developing programs in a computer and to give the effective outputs by using the experience

Selenoprotein P (SELENOP) acts as a crucial mediator, distributing selenium from the liver to other tissues within the body. Despite its established role in selenium metabolism, the specific functions of SELENOP in the development of liver cancer remain enigmatic. This study aims to unravel SELENOP's associations in hepatocellular carcinoma (HCC) by scrutinizing its expression in correlation with disease characteristics and investigating links to hormonal and lipid/triglyceride metabolism biomarkers as well as its potential as a prognosticator for overall survival and predictor of hypoxia. SELENOP mRNA expression was analyzed in 372 HCC patients sourced from The Cancer Genome Atlas (TCGA), utilizing statistical methodologies in R programming and machine learning techniques in Python. SELENOP expression significantly varied across HCC grades (p < 0.000001) and among racial groups (p = 0.0246), with lower levels in higher grades and Asian individuals, respectively. Gender significantly influenced SELENOP expression (p < 0.000001), with females showing lower altered expression compared to males. Notably, the Spearman correlation revealed strong positive connections of SELENOP with hormonal markers (AR, ESR1, THRB) and key lipid/triglyceride metabolism markers (PPARA, APOC3, APOA5). Regarding prognosis, SELENOP showed a significant association with overall survival (p = 0.0142) but explained only a limited proportion of variability (~10%). Machine learning suggested its potential as a predictive biomarker for hypoxia, explaining approximately 18.89% of the variance in hypoxia scores. Future directions include validating SELENOP's prognostic and diagnostic value in serum for personalized HCC treatment. Large-scale prospective studies correlating serum SELENOP levels with patient outcomes are essential, along with integrating them with clinical parameters for enhanced prognostic accuracy and tailored therapeutic strategies.

The relevance of the research is due to the growing need to automate the processes of text analysis and correction, in particular for Ukrainian-language content, which is characterized by a wealth of morphological and syntactic structure. Due to the wide range of errors that can occur in texts, from spelling to contextual, there is an urgent need to create systems that can accurately identify errors and offer their correct corrections. The specificity of the Ukrainian language, including its grammatical complexity and multifacetedness, requires the adaptation of machine learning models to local features. The purpose of the research is to develop a mathematical model of a decision support system for identifying and correcting errors in Ukrainian-language texts. The task includes both the formalization and mathematical description of the text processing process, and the construction of a model with an orientation to the tasks of classification and text generation. Special attention is paid to the effective consideration of structural features specific to the Ukrainian language in order to increase the accuracy and productivity of the system. The research method is based on the construction of a mathematical model of error correction, which is presented as a context-aware text generation problem. The study used statistical methods and machine learning approaches. Special attention is paid to the formation of a training sample, which combines texts with real and artificial errors to ensure a balanced learning process. The correction module includes generation mechanisms based on contextual models capable of predicting the correct correction for erroneous tokens. Approaches to text vectorization are mathe- matically substantiated, taking into account the peculiarities of the morphology and syntax of the Ukrainian language. The constructed model is a universal basis for creating intelligent systems for automatic editing of Ukrainian-language text. As a result of the research, approaches to building an error correction model in Ukrainian-language texts are formulated and mathematically substantiated. The main result was the creation of an integrated system that uses contextual information to ensure high accuracy of error recognition and correction. The applied mathematical methods include probabilistic approaches and vector representation of tokens, which allows adapting the system to the peculiarities of the Ukrainian language with its high morphological and syntactic complexity. The formed basis of the model creates opportunities for scaling and further use in practical tasks, such as automatic text editing or improving the quality of content in the Ukrainian-speaking environment.