A Review on Large Language Models: Architectures, Applications, Taxonomies, Open Issues and Challenges

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 26839 - 26874

Published: Jan. 1, 2024

Large Language Models (LLMs) recently demonstrated extraordinary capability, including natural language processing (NLP), translation, text generation, question answering, etc. Moreover, LLMs are a new and essential part of computerized processing, having the ability to understand complex verbal patterns generate coherent appropriate replies for situation. Though this success has prompted substantial increase in research contributions, rapid growth made it difficult overall impact these improvements. Since lot on is coming out quickly, getting tough get an overview all them short note. Consequently, community would benefit from but thorough review recent changes area. This article thoroughly overviews LLMs, their history, architectures, transformers, resources, training methods, applications, impacts, challenges, paper begins by discussing fundamental concepts with its traditional pipeline phase. It then provides existing works, history evolution over time, architecture transformers different resources methods that have been used train them. also datasets utilized studies. After that, discusses wide range applications biomedical healthcare, education, social, business, agriculture. illustrates how create society shape future AI they can be solve real-world problems. Then explores open issues challenges deploying scenario. Our aims help practitioners, researchers, experts pre-trained goals.

Language: Английский

---

Artificial Intelligence in Healthcare: Review, Ethics, Trust Challenges & Future Research Directions

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 120, P. 105894 - 105894

Published: Jan. 28, 2023

Language: Английский

---

Deep learning techniques for detection and prediction of pandemic diseases: a systematic literature review

Multimedia Tools and Applications, Journal Year: 2023, Volume and Issue: 83(2), P. 5893 - 5927

Published: May 29, 2023

Abstract Deep learning (DL) is becoming a fast-growing field in the medical domain and it helps timely detection of any infectious disease (IDs) essential to management diseases prediction future occurrences. Many scientists scholars have implemented DL techniques for pandemics, IDs other healthcare-related purposes, these outcomes are with various limitations research gaps. For purpose achieving an accurate, efficient less complicated DL-based system therefore, this study carried out systematic literature review (SLR) on pandemics using techniques. The survey anchored by four objectives state-of-the-art forty-five papers seven hundred ninety retrieved from different scholarly databases was analyze evaluate trend application areas pandemics. This used tables graphs extracted related articles online repositories analysis showed that good tool pandemic prediction. Scopus Web Science given attention current because they contain suitable scientific findings subject area. Finally, presents forty-four (44) studies technique performances. challenges identified include low performance model due computational complexities, improper labeling absence high-quality dataset among others. suggests possible solutions such as development improved or reduction output layer architecture pandemic-prone considerations.

Language: Английский

---

Federated Learning for Healthcare: A Comprehensive Review

Published: Feb. 9, 2024

Recent advancements in deep learning for healthcare and computer-aided laboratory services have sparked a renewed interest making medical data more accessible. Elevating the quality of delivering improved patient care necessitates knowledge base rooted data-driven insights. Deep models proven to excel this regard, as they are specifically designed embrace approach. These thrive on exposure larger datasets, which enables them continuously improve their performance. However, organizations strive aggregate clinical records onto central servers construct robust models, concerns surrounding privacy, ownership, legal restrictions emerged. Safeguarding sensitive while harnessing collective from multiple centers is challenging balancing act. One promising approach address these use privacy-preserving techniques that allow utilization without compromising security. Federated (FL) technique has emerged enable deployment large machine trained across necessity sharing information. In article, we present most recent findings derived systematic literature review focusing application federated settings. This offers insights into current state research practical implementations FL within domain. By leveraging learning, institutions can harness power upholding privacy security standards, ultimately leading effective solutions.

Language: Английский

---

Artificial Intelligence for Neuroimaging in Pediatric Cancer

Cancers, Journal Year: 2025, Volume and Issue: 17(4), P. 622 - 622

Published: Feb. 12, 2025

Background/Objectives: Artificial intelligence (AI) is transforming neuroimaging by enhancing diagnostic precision and treatment planning. However, its applications in pediatric cancer remain limited. This review assesses the current state, potential applications, challenges of AI for cancer, emphasizing unique needs population. Methods: A comprehensive literature was conducted, focusing on AI’s impact through accelerated image acquisition, reduced radiation, improved tumor detection. Key methods include convolutional neural networks segmentation, radiomics characterization, several tools functional imaging. Challenges such as limited datasets, developmental variability, ethical concerns, need explainable models were analyzed. Results: has shown significant to improve imaging quality, reduce scan times, enhance accuracy neuroimaging, resulting segmentation outcome prediction treatment. progress hindered scarcity issues with data sharing, implications applying vulnerable populations. Conclusions: To overcome limitations, future research should focus building robust fostering multi-institutional collaborations developing interpretable that align clinical practice standards. These efforts are essential harnessing full improving outcomes children cancer.

Language: Английский

---

Improvement of Classification Accuracy in Machine Learning Algorithm by Hyper-Parameter Optimization

Published: Nov. 1, 2023

The manual optimization of hyperparameters is a straightforward and well-known approach, but it not scalable, particularly when there are several settings options. In nearly every area daily life, machine learning offers more logical guidance than humans can. It has already been noted in the literature that correct Hyper-Parameter significant impact on algorithm's performance. Manual search one method for performing optimization, however takes lot time. Some common techniques used hyperparameter include grid search, random procedure. main model training structural hyper-parameters introduced first part, along with their significance approaches defining value range. research then concentrates applicability, examining effectiveness accuracy, forest ensemble algorithm. this study, we present novel approach enhancing Random Forest using Parkinson's Disease Data Set. Accuracy, precision, recall F1 score were taken into account while comparing performances each these strategies.

Language: Английский

---

A Review on Large Language Models: Architectures, Applications, Taxonomies, Open Issues and Challenges

Published: Sept. 27, 2023

Large Language Models (LLMs) recently demonstrated extraordinary capability in various natural language processing (NLP) tasks including translation, text generation, question answering, etc.Moreover, LLMs are new and essential part of computerized processing, having the ability to understand complex verbal patterns generate coherent appropriate replies a given context.Though this success has prompted substantial increase research contributions, rapid growth made it difficult overall impact these improvements.Since plethora on have been appeared within short time, is quite impossible track all get an overview current state area.Consequently, community would benefit from but thorough review recent changes area.This article thoroughly overviews LLMs, their history, architectures, transformers, resources, training methods, applications, impacts, challenges, etc.This paper begins by discussing fundamental concepts with its traditional pipeline phase.Then provides existing works, history evolution over architecture transformers different resources methods that used train them.The also demonstrates datasets utilized studies.After that, discusses wide range applications biomedical healthcare, education, social, business, agriculture.The study illustrates how create society shape future AI they can be solve real-world problems.Finally, explores open issues challenges deploy scenario.Our aims help practitioners, researchers, experts pre-trained goals.

Language: Английский

---

Generative artificial intelligence: synthetic datasets in dentistry

BDJ Open, Journal Year: 2024, Volume and Issue: 10(1)

Published: March 1, 2024

Abstract Introduction Artificial Intelligence (AI) algorithms, particularly Deep Learning (DL) models are known to be data intensive. This has increased the demand for digital in all domains of healthcare, including dentistry. The main hindrance progress AI is access diverse datasets which train DL ensuring optimal performance, comparable subject experts. However, administration these traditionally acquired challenging due privacy regulations and extensive manual annotation required by Biases such as ethical, socioeconomic class imbalances also incorporated during curation datasets, limiting their overall generalizability. These challenges prevent accrual at a larger scale training models. Methods Generative techniques can useful production Synthetic Datasets (SDs) that overcome issues affecting datasets. Variational autoencoders, generative adversarial networks diffusion have been used generate SDs. following text review operations. It discusses chances SDs with potential solutions will improve understanding healthcare professionals working research. Conclusion customized need researchers produced robust models, having trained on dataset applicable dissemination across countries. there limitations associated better understood, attempts made those concerns prior widespread use.

Language: Английский

---

Artificial intelligence methods for modeling gasification of waste biomass: a review

Environmental Monitoring and Assessment, Journal Year: 2024, Volume and Issue: 196(3)

Published: Feb. 26, 2024

Language: Английский

---

Sexual health in the era of artificial intelligence: a scoping review of the literature

Sexual Medicine Reviews, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

Abstract Introduction Artificial Intelligence (AI) has witnessed significant growth in the field of medicine, leveraging machine learning, artificial neuron networks, and large language models. These technologies are effective disease diagnosis, education, prevention, while raising ethical concerns potential challenges. However, their utility sexual medicine remains relatively unexplored. Objective We aim to provide a comprehensive summary status AI medicine. Methods A search was conducted using MeSH keywords, including "artificial intelligence," "sexual medicine," health," "machine learning." Two investigators screened articles for eligibility within PubMed MEDLINE databases, with conflicts resolved by third reviewer. Articles English that reported on health were included. total 69 full-text systematically analyzed based predefined inclusion criteria. Data extraction included information article characteristics, study design, assessment methods, outcomes. Results The initial yielded 905 relevant Upon assessing full texts 121 eligibility, 52 studies unrelated excluded, resulting systematic review. analysis revealed AI's accuracy preventing, diagnosing, decision-making sexually transmitted diseases. also demonstrated ability diagnose offer precise treatment plans male female dysfunction infertility, accurately predict sex from bone teeth imaging, correctly orientation relationship issues. emerged as promising modality implications future Conclusions Further research is essential unlock presents advantages such accessibility, user-friendliness, confidentiality, preferred source information. it still lags human healthcare providers terms compassion clinical expertise.

Language: Английский

---