Weighted Average Ensemble Deep Learning Model for Stratification of Brain Tumor in MRI Images

Diagnostics, Год журнала: 2023, Номер 13(7), С. 1320 - 1320

Опубликована: Апрель 2, 2023

Brain tumor diagnosis at an early stage can improve the chances of successful treatment and better patient outcomes. In biomedical industry, non-invasive diagnostic procedures, such as magnetic resonance imaging (MRI), be used to diagnose brain tumors. Deep learning, a type artificial intelligence, analyze MRI images in matter seconds, reducing time it takes for potentially improving Furthermore, ensemble model help increase accuracy classification by combining strengths multiple models compensating their individual weaknesses. Therefore, this research, weighted average deep learning is proposed For model, three different feature spaces are taken from transfer VGG19 Convolution Neural Network (CNN) without augmentation, CNN with augmentation. These ensembled best combination weights, i.e., weight1, weight2, weight3 using grid search. The dataset simulation Cancer Genome Atlas (TCGA), having lower-grade glioma collection 3929 110 patients. helps reduce overfitting that have learned aspects data. outperforms detecting tumors terms accuracy, precision, F1-score. act second opinion tool radiologists brain.

Язык: Английский

---

A systematic review of hyperparameter optimization techniques in Convolutional Neural Networks

Decision Analytics Journal, Год журнала: 2024, Номер 11, С. 100470 - 100470

Опубликована: Апрель 24, 2024

Convolutional Neural Network (CNN) is a prevalent topic in deep learning (DL) research for their architectural advantages. CNN relies heavily on hyperparameter configurations, and manually tuning these hyperparameters can be time-consuming researchers, therefore we need efficient optimization techniques. In this systematic review, explore range of well used algorithms, including metaheuristic, statistical, sequential, numerical approaches, to fine-tune hyperparameters. Our offers an exhaustive categorization (HPO) algorithms investigates the fundamental concepts CNN, explaining role variants. Furthermore, literature review HPO employing above mentioned undertaken. A comparative analysis conducted based strategies, error evaluation accuracy results across various datasets assess efficacy methods. addition addressing current challenges HPO, our illuminates unresolved issues field. By providing insightful evaluations merits demerits objective assist researchers determining suitable method particular problem dataset. highlighting future directions synthesizing diversified knowledge, survey contributes significantly ongoing development optimization.

Язык: Английский

---

Character Recognition Technique Implementation for Complicated Deteriorated Scene

2021 5th International Conference on Information Systems and Computer Networks (ISCON), Год журнала: 2023, Номер unknown, С. 1 - 4

Опубликована: Март 3, 2023

There has been an increase in interest digitizing and preserving old books papers the last few years. The quick advancement of data innovation Internet's spread also contributed to enormous volume image video data. texts that are included assist us analyzing them utilized for indexing, archiving, retrieval. Different noises, such as Gaussian noise, salt pepper speckle etc., can readily damage image. Several filtering algorithms, including filter, mean median employed eliminate these various noises from images. This article analyses impact several pre-processing approaches, thresholding, morphology, blurring procedures, maximise text extraction strategies. experiment's findings demonstrate approaches unquestionably improve document's structural visual quality.

Язык: Английский

---

A hybrid chromaticity-morphological machine learning model to overcome the limit of detecting newcastle disease in experimentally infected chicken within 36 h

Computers and Electronics in Agriculture, Год журнала: 2025, Номер 234, С. 110248 - 110248

Опубликована: Март 19, 2025

Язык: Английский

---

Optimizing Convolutional Neural Networks: A Comprehensive Review of Hyperparameter Tuning Through Metaheuristic Algorithms

Archives of Computational Methods in Engineering, Год журнала: 2025, Номер unknown

Опубликована: Май 20, 2025

Язык: Английский

---

Segmentation and Classification of Encephalon Tumor by Applying Improved Fast and Robust FCM Algorithm with PSO-Based ELM Technique

Computational Intelligence and Neuroscience, Год журнала: 2022, Номер 2022, С. 1 - 9

Опубликована: Июль 31, 2022

Nowadays, so many people are living in world. If living, then the diseases also increasing day by due to adulterated and chemical content food. The may suffer either from a small disease such as cold cough or big cancer. In this work, we have discussed on encephalon tumor cancer which is problem nowadays. will consider about whole world, there deficiency of clinical experts doctors compared affected person. So, here, used an automatic classification help particle swarm optimization (PSO)-based extreme learning machine (ELM) technique with segmentation process improved fast robust fuzzy C mean (IFRFCM) algorithm most commonly feature reduction method gray level co-occurrence matrix (GLCM) that helpful experts. Here, BraTs ("Multimodal Brain Tumor Segmentation Challenge 2020") dataset for both training testing purpose. It has been monitored our system given better accuracy approximation 99.47% can be observed good outcome.

Язык: Английский

---

Explicability of Artificial Intelligence in Healthcare 5.0

Опубликована: Янв. 27, 2023

Marvin Minsky, the father of Artificial intelligence (AI), defined AI as machines that are smarter than humans and can-do tasks cannot easily do. has grown at a rapid rate never before seen in variety industries. Different types already used by insurance companies, service providers companies. Now, with addition AI, healthcare sector is also emerging. Increased availability medical data advances analytical techniques causing paradigm shift healthcare. The literature reveals multiple applications for services an incompletely covered body research. authors explored current situation technology was examined, to their long-term prospects. As result, goal use bibliometric method analyze dynamics interconnection between digital health approaches while taking into account responsible ethical elements scientific output throughout time.

Язык: Английский

---

Machine Learning and Its Application in Educational Area

Опубликована: Март 8, 2024

Machine learning (ML) is used for advancement of various fields which also applicable to education system will change and teaching methods fundamentally. As educational institutions gather a sizable amount student data, this information can be further narrow down the elements that changed improve likelihood students succeed. ML utilized by educators like retention better grading systems results. Development new insights done using ML. This study discusses how we use in sector tackle problems from students' teachers' perspective them future research on topic.

Язык: Английский

---

An Investigation on CNN-based Lung Cancer Prediction Method

2022 International Conference on Computer Communication and Informatics (ICCCI), Год журнала: 2023, Номер unknown

Опубликована: Янв. 23, 2023

Diagnosis of any disease is done only after detection the type and intensity disease. Chronic pulmonary such as Obstructive Pulmonary Disease (COPD), lung cancer, cystic fibrosis needs to be detected in initial stages get proper treatment. However, early diagnosis cancer case lungs not very easy due presence no recognizable symptoms until much more adverse situations. The proposed work aims at detecting using Convolutional Neural Network (CNN) regarding dataset compiled from Kaggle Domain. Various parameters are used determine performance method, namely, precision, recall F1-score. Experimental results concluded that, CNN-based approach resulted significantly improved when compared few existing methods. method has 94% average 93.6% 94.6% F1-score, respectively.

Язык: Английский

---

Synthetic Image Generation With a Fine-Tuned Latent Diffusion Model for Organ on Chip Cell Image Classification

Опубликована: Сен. 20, 2023

Augmentation of the datasets authentic microscopy images with synthetic is a promising solution to problem limited availability biomedical data for training deep neural network (DNN) based classifiers. In present study, we use text-to-image latent stable diffusion model fine-tuned by means low-rank adaptation (LoRA) augment small dataset organ on chip cells. While resulting appear quite similar which was performed, find that neither EfficientNetB7 DNN solely nor augmentation real-world different proportions (10, 25, 50, and 75 percent) these leads improvement accuracy model. The findings our study suggest further exploration options needed fully capacity models synthesis images.

Язык: Английский

---