Ayhan ARISOY

Dr. Ayhan ARISOY

Bucak Faculty of Computer and Informatics, Mehmet Akif Ersoy University, Burdur, Turkey (Türkiye)
orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 March 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
MAMerve Varol Arısoy
Merve Varol Arısoy
Ayhan ARISOYAyhan ARISOY
Ayhan ARISOY
IUIlhan Uysal
Ilhan Uysal
This study introduces the Medical Vision Attention Generation (MedVAG) model, a novel framework designed to facilitate the automated generation of medical reports. MedVAG integrates Vision Transformer (ViT)-based visual feature extraction and GPT-2 language modeling, enhanced by graph-based feature fusion and multiple attention mechanisms (co-attention, cross-attention, memory-guided attention), to ensure semantic coherence and diagnostic accuracy. Evaluated on IU X-Ray and COV-CTR datasets, the model achieved state-of-the-art performance across natural language generation metrics (BLEU, METEOR, ROUGE, CIDEr) and clinical effectiveness measures. Ablation studies highlighted the critical role of attention mechanisms and feature fusion in aligning visual and textual features. MedVAG demonstrates strong potential as an assistive technology, aiming to support radiologists by reducing workload and enhancing diagnostic accuracy.

8 February 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
MAMerve Varol Arısoy
Merve Varol Arısoy
MYMehmet Ali Yalçınkaya
Mehmet Ali Yalçınkaya
RGRemzi GÜRFİDAN
Remzi GÜRFİDAN
Ayhan ARISOYAyhan ARISOY
Ayhan ARISOY
The aim of this study is to detect expressions of violence against healthcare workers using natural language processing techniques. Experiments on various NLP models have shown that violent expressions can be successfully classified using textual data. The RAG-ECE model performed the best in this study with a 97.97% accuracy rate and a 97.67% F1 score. The model provided a strong balancing performance in the “no violence” class with 97.71% precision and 97.67% recall rates. In the “violence present” class, it reached 97.67% accuracy and was evaluated as a reliable classifier with both low false positive (3.92%) and low false negative (2.78%) rates. In addition to RAG-ECE, the GPT model provided a milder alternative with 96.19% accuracy and a 96.26% F1 score. The study also compared the performances of other models, such as GPT, BERT, SVM, and NB, and stated that they are considered suitable alternatives due to their low computational costs, especially in small- and medium-sized datasets. The findings of the study show that NLP-based systems offer an effective solution for the early detection and prevention of expressions of violence against healthcare workers.

30 October 2024 in Yalvaç Akademi Dergisi
Verified
Journal Article
Closed access
Ayhan ARISOYAyhan ARISOY
Ayhan ARISOY
Natural language processing (NLP) is the general name for the methods and algorithms developed for computers to understand, interpret and produce human language. NLP plays a critical role in many fields, from social media analyses to customer service, from language translation to healthcare. This paper provides a comprehensive overview of the basic concepts of NLP, popular algorithms and models, performance comparisons, and various application areas. Key concepts of NLP include language models, tokenisation, lemmatisation, stemming, POS tagging, NER and syntactic parsing. These concepts are critical for processing, analysing and making sense of texts. Language models include popular methods such as N-gram, Word2Vec, GloVe and BERT. NLP algorithms are classified as rule-based methods, machine learning methods and deep learning methods. Rule-based methods are based on grammatical rules, while machine learning methods work on the principle of learning from data. Deep learning methods, on the other hand, achieve high accuracy results by using large datasets and powerful computational resources. In the performance comparison section, it is stated that the algorithms are evaluated with metrics such as accuracy, precision, recall and F1 score. Advanced models such as BERT and GPT-3 show superior performance in many NLP tasks. In conclusion, the field of NLP is rapidly evolving, with significant advancements anticipated in several key areas. These include the creation of more effective and efficient models, efforts to reduce biases, enhanced privacy protection, the growth of multilingual and cross-cultural models, and the development of explainable artificial intelligence techniques. This paper provides a comprehensive overview to understand the current status and future directions of NLP technologies.

30 September 2024 in Bilge International Journal of Science and Technology Research
Verified
Journal Article
Open access
PDF file
Ayhan ARISOYAyhan ARISOY
Ayhan ARISOY
EAENES AÇIKGÖZOĞLU
ENES AÇIKGÖZOĞLU
Nowadays, the sustainability of agriculture and food security have an increasing importance on soil fertility. Soil fertility is defined as the capacity of a land to grow crops and its potential crop productivity. However, factors such as increasing population, climate change, land use changes and environmental pollution threaten soil fertility. These threats can result in problems such as erosion, soil salinisation and organic matter depletion. Soil fertility is critical for the long-term health of agriculture and food security. Artificial intelligence techniques used to determine and manage soil fertility analyse the minerals present in the soil as well as other factors. These analyses assess the amount of minerals present in the soil, the availability of nutrients and important parameters such as pH. This information guides farmers in selecting the most appropriate crops. Furthermore, the integration of Internet of Things (IoT) technologies allows real-time monitoring of minerals and nutrients in the soil and optimising irrigation and fertilisation processes based on this data. These developments have the potential to improve soil fertility management and increase agricultural productivity.

25 August 2025 in SDÜ Fen Bilimleri Enstitüsü Dergisi
Verified
Journal Article
Open access
PDF file
Ayhan ARISOYAyhan ARISOY
Ayhan ARISOY
Radiology reports contain clinically critical yet complex information that can overwhelm physicians when available in a raw narrative form. To address this challenge, this study proposes a Retrieval-Augmented Generation (RAG) approach to medical report summarization by integrating a FAISS-based semantic search engine with a small T5 generative model. The methodology uses Sentence-BERT to embed the “Findings” section of radiology reports from the MIMIC-III dataset, which are then indexed using FAISS to retrieve semantically similar cases. In this approach, The retrieved context has been merged with the original 'Finding' section and subsequently fed into the T5 model to produce the respective 'Impression' summary. Experimental results demonstrate the model’s effectiveness in terms of both lexical fidelity and semantic consistency, achieving ROUGE-1 of 0.5299, ROUGE-2 of 0.4206, METEOR of 0.5018, and compression ratio of 0.9213. Domain-specific semantic matching was validated using FactCC (0.5463), CheXpert Label Agreement (0.6194), and Medical Concept Overlap (0.5552) scores. The model demonstrated stable convergence over seven epochs without overfitting, as evidenced by the steadily decreasing validation loss. Qualitative examples show that the model generates smooth and clinically consistent summaries, although occasional factual hallucinations indicate areas for improvement. Overall, the proposed FAISS+T5 process addresses key limitations of traditional summarization methods by integrating contextual retrieval and generation. This framework offers a scalable, interpretable, and domain-specific solution for summarizing clinical texts and shows promise for implementation in real-world decision support systems.

8 August 2026 in The Journal of Supercomputing
Verified
Journal Article
Closed access
PDF file
Ayhan ARISOYAyhan ARISOY
Ayhan ARISOY
Ecir Uğur KüçüksilleEcir Uğur Küçüksille
Ecir Uğur Küçüksille