Breast Cancer Diagnosis Using YOLO-Based Multiscale Parallel CNN and Flattened Threshold Swish

Applied Sciences, Год журнала: 2024, Номер 14(7), С. 2680 - 2680

Опубликована: Март 22, 2024

In the field of biomedical imaging, use Convolutional Neural Networks (CNNs) has achieved impressive success. Additionally, detection and pathological classification breast masses creates significant challenges. Traditional mammogram screening, conducted by healthcare professionals, is often exhausting, costly, prone to errors. To address these issues, this research proposes an end-to-end Computer-Aided Diagnosis (CAD) system utilizing ‘You Only Look Once’ (YOLO) architecture. The proposed framework begins enhancing digital mammograms using Contrast Limited Adaptive Histogram Equalization (CLAHE) technique. Then, features are extracted CNN, leveraging multiscale parallel feature extraction capabilities while incorporating DenseNet InceptionNet architectures. combat ‘dead neuron’ problem, CNN architecture utilizes ‘Flatten Threshold Swish’ (FTS) activation function. YOLO loss function been enhanced effectively handle lesion scale variation in mammograms. was thoroughly tested on two publicly available benchmarks: INbreast CBIS-DDSM. It accuracy 98.72% for cancer dataset a mean Average Precision (mAP) 91.15% utilized only 11.33 million parameters training. These results highlight framework’s ability revolutionize vision-based diagnosis.

Язык: Английский

---

Determination of platinum-resistance of women with ovarian cancer by FTIR spectroscopy combined with multivariate analyses and machine learning methods

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Окт. 22, 2024

Patients with high-grade ovarian cancer have a poor prognosis, thus effective treatment remains an unmet medical issue of high importance. Moreover, finding the reason for resistance to cisplatin is crucial task improvement anti-cancer drugs. In this study, we showed first time chemical difference in serum collected from platinum-resistance and platinum-sensitive women suffering using Fourier Transform InfraRed (FTIR) spectroscopy followed by data analysis Principal Component Analysis (PCA), Hierarchical (HCA) 4 different machine learning algorithms. Obtained results shift PO2-symmetric vibrations, amide III II were observed on FTIR spectrum comparison women. Furthermore, PCA clearly demonstrated most important role I differentiation addition, algorithms wavenumber at 1631 cm-1(amide I) 2993 cm-1 (asymmetric stretching CH3 vibrations). The accuracy obtained was above 92%. Summarizing, can be used detection phenomena.

Язык: Английский

---

Heart disease detection system based on ECG and PCG signals with the aid of GKVDLNN classifier

Multimedia Tools and Applications, Год журнала: 2023, Номер 83(10), С. 30587 - 30612

Опубликована: Сен. 7, 2023

Язык: Английский

---

Enhancing security in smart healthcare systems: Using intelligent edge computing with a novel Salp Swarm Optimization and radial basis neural network algorithm

Heliyon, Год журнала: 2024, Номер 10(13), С. e33792 - e33792

Опубликована: Июнь 28, 2024

A smart healthcare system (SHS) is a health service that employs advanced technologies such as wearable devices, the Internet of Things (IoT), and mobile internet to dynamically access information connect people institutions related healthcare, thereby actively managing responding medical ecosystem needs. Edge computing (EC) plays significant role in SHS it enables real-time data processing analysis at source, which reduces latency improves intervention speed. However, integration patient information, including electronic records (EHRs), into framework induces security privacy concerns. To address these issues, an intelligent EC was proposed this study. The objective study accurately identify threats ensure secure transmission environment. leverages effectiveness Salp Swarm Optimization Radial Basis Functional Neural Network (SS-RBFN) for enhancing privacy. methodology commences with collection then pre-processed consistency quality database further analysis. Subsequently, SS-RBFN algorithm trained using distinguish between normal malicious streams accurately, offering continuous monitoring Additionally, Rivest-Shamir-Adelman (RSA) approach applied safeguard against during cloud storage. model validated IoT-based available Kaggle, experimental results demonstrated achieved 99.87 % accuracy, 99.76 precision, 99.49 f-measure, 98.99 recall, 97.37 throughput, 1.2s latency. Furthermore, by were compared existing models validate its security.

Язык: Английский

---

Enhancing breast cancer treatment selection through 2TLIVq-ROFS-based multi-attribute group decision making

Frontiers in Artificial Intelligence, Год журнала: 2024, Номер 7

Опубликована: Июнь 3, 2024

Introduction Breast cancer is an extremely common and potentially fatal illness that impacts millions of women worldwide. Multiple criteria inclinations must be taken into account when selecting the optimal treatment option for each patient. Methods The selection breast treatments can modeled as a multi-attribute group decision-making (MAGDM) problem, in which experts evaluate rank alternative based on multiple attributes. MAGDM methods aid enhancing quality efficacy decisions. For this purpose, we introduce concept 2-tuple linguistic interval-valued q -rung orthopair fuzzy set (2TLIV -ROFS), new development theory incorporates characteristics (IV -ROFS) terms. It express quantitative qualitative aspects uncertain information, well decision-makers' level satisfaction dissatisfaction. Results Then, 2TLIV -ROF weighted average -ROFWA) operator geometric -ROFWJ) are introduced two aggregation operators. In addition, border approximation area comparison (MABAC) method extended to solve problem with information. Discussion To demonstrate applicability suggested model, case study presented. results computations show model able handle imprecision subjectivity complicated scenarios opens research scholars.

Язык: Английский

---

Breast Cancer Classification Using Customized Convolution Neural Network

Опубликована: Май 9, 2024

Язык: Английский

---