Segmenting and classifying lung diseases with M-Segnet and Hybrid Squeezenet-CNN architecture on CT images DOI Creative Commons

Syed Mohammed Shafi,

C. Sathiya Kumar

PLoS ONE, Год журнала: 2024, Номер 19(5), С. e0302507 - e0302507

Опубликована: Май 16, 2024

Diagnosing lung diseases accurately and promptly is essential for effectively managing this significant public health challenge on a global scale. This paper introduces new framework called Modified Segnet-based Lung Disease Segmentation Severity Classification (MSLDSSC). The MSLDSSC model comprises four phases: "preprocessing, segmentation, feature extraction, classification." Initially, the input image undergoes preprocessing using an improved Wiener filter technique. technique estimates power spectral density of noisy original images computes SNR assisted by PSNR to evaluate quality. Next, preprocessed identify separate RoI from background objects in image. We employ Segnet mechanism that utilizes proposed hard tanh-Softplus activation function effective Segmentation. Following Segmentation, features such as MLDN, entropy with MRELBP, shape features, deep are extracted. extraction phase, retrieved set into hybrid severity classification model. two classifiers: SDPA-Squeezenet DCNN. These classifiers train classify level diseases.

Язык: Английский

Towards a Secure and Sustainable Internet of Medical Things (IoMT): Requirements, Design Challenges, Security Techniques, and Future Trends DOI Open Access
Bharat Bhushan,

Avinash Kumar,

Ambuj Kumar Agarwal

и другие.

Sustainability, Год журнала: 2023, Номер 15(7), С. 6177 - 6177

Опубликована: Апрель 3, 2023

Recent advances in machine-to-machine (M2M) communications, mini-hardware manufacturing, and micro computing have led to the development of Internet Things (IoT). The IoT is integrated with medical devices order enable better treatment, cost-effective solutions, improved patient monitoring, enhanced personalized healthcare. This has more complex heterogeneous Medical (IoMT) systems that their own operating protocols. Even though such pervasive low-cost sensing can bring about enormous changes healthcare sector, these are prone numerous security privacy issues. Security thus a major challenge critical systems, one inhibits widespread adoption. However, significant inroads been made by on-going research, which powers IoMT applications incorporating prevalent measures. In this regard, paper highlights significance implementing key measures, essential aspects make it useful for interconnecting various internal external working domains presents state-of-the-art techniques securing terms data transmission, collection, storage. Furthermore, also explores requirements, inherent design challenges, could secure sustainable. Finally, gives panoramic view current status research field outlines some future directions area.

Язык: Английский

Процитировано

58
AML‐Net: Attention‐based multi‐scale lightweight model for brain tumour segmentation in internet of medical things DOI Creative Commons
Muhammad Zeeshan Aslam, Basit Raza, Muhammad Faheem

и другие.

CAAI Transactions on Intelligence Technology, Год журнала: 2024, Номер unknown

Опубликована: Янв. 17, 2024

Abstract Brain tumour segmentation employing MRI images is important for disease diagnosis, monitoring, and treatment planning. Till now, many encoder‐decoder architectures have been developed this purpose, with U‐Net being the most extensively utilised. However, these require a lot of parameters to train semantic gap. Some work tried make lightweight model do channel pruning that made small receptive field which compromised accuracy. The authors propose an attention‐based multi‐scale called AML‐Net in Internet Medical Things overcome above issues. This consists three are trained different scale input along previously learned features diminish loss. Moreover, designed attention module replaced traditional skip connection. For module, six experiments were conducted, from dilated convolution spatial performed well. has convolutions relatively large followed by extract global context encoder low‐level features. Then fine combined decoder's same layer high‐level perform experiment on low‐grade‐glioma dataset provided Cancer Genome Atlas at least Fluid‐Attenuated Inversion Recovery modality. proposed 1/43.4, 1/30.3, 1/28.5, 1/20.2 1/16.7 fewer than Z‐Net, U‐Net, Double BCDU‐Net CU‐Net respectively. authors’ gives results IoU = 0.834, F 1‐score 0.909 sensitivity 0.939, greater CU‐Net, RCA‐IUnet PMED‐Net.

Язык: Английский

Процитировано

9
IoT-Enabled Secure and Intelligent Smart Healthcare DOI
Wasswa Shafik

Advances in computational intelligence and robotics book series, Год журнала: 2024, Номер unknown, С. 308 - 333

Опубликована: Апрель 1, 2024

This study examines the complex array of impediments and potential advantages internet things (IoT)-enabled secure intelligent smart healthcare devices (IESISHDs) associated with shift towards enabling cities, motivated by pressing necessity to address climate change promote sustaining systems. looks at technological, economic, social problems that need be solved in order make cities smarter IoT. It does this reading a lot scholarly sources. Most stupendously, it emphasizes environmentally sustainable merits, for economic growth, improvements societal well-being can arise from transition. further depicts selected case studies demonstrate empirical evidence provide policy recommendations. The paradigm is assist governments other stakeholders effectively managing human-associated challenges attain maximum value an innovative future guarantees worldwide prosperity ecological welfare.

Язык: Английский

Процитировано

9
Artificial Intelligence-Enabled Internet of Medical Things (AIoMT) in Modern Healthcare Practices DOI
Wasswa Shafik

Advances in medical technologies and clinical practice book series, Год журнала: 2024, Номер unknown, С. 42 - 69

Опубликована: Июнь 7, 2024

The integration of artificial intelligence (AI), the internet things (IoT), with medical devices avails recent development in sector, specifically digital health, referred to as (IoMT). AIoMT combines technologies like body movement detection, sleep monitoring, and rehab assessment, simplifying healthcare offering personalized experiences. By leveraging AI, big data, mobile internet, cloud computing, microelectronics, patient data is efficiently processed, enhancing healthcare's efficiency personalization. During pandemic, AI applications saved lives by streamlining analysis. This chapter explores wearable electronics sensor architecture addresses challenges security, aiming elevate standards. It also outlines future research opportunities AIoMT.

Язык: Английский

Процитировано

7
IoMT Future Trends and Challenges DOI
Wasswa Shafik

Advances in healthcare information systems and administration book series, Год журнала: 2024, Номер unknown, С. 348 - 370

Опубликована: Май 17, 2024

The healthcare industry is transforming significantly due to the rapid emergence of internet medical things (IoMT). integration cutting-edge technologies facilitates this paradigm shift. A new age system optimization and patient care being ushered in. This study provides a comprehensive overview future trends open issues in adopting IoMTs. It explores current status IoMT forecasts its evolution. examines policy regulatory ramifications essential ethical data privacy aspects. More still elucidates urgent security, interoperability, scalability difficulties while underscoring imperative for collaborative efforts standards within industry. affords insights research by presenting set unanswered inquiries corresponding possible implications, accompanied relevant cases. Finally, it emphasizes significant impact can have on availing lightweight digital trust architectures.

Язык: Английский

Процитировано

5
PulmonU-Net: a semantic lung disease segmentation model leveraging the benefit of multiscale feature concatenation and leaky ReLU DOI Creative Commons

H. Mary Shyni,

E. Chitra

Automatika, Год журнала: 2024, Номер 65(2), С. 641 - 651

Опубликована: Фев. 13, 2024

Pulmonary diseases impact lung functionality and can cause health complications. X-ray imaging is an initial diagnostic approach for evaluating conditions. Manual segmentation of infections from X-rays time-consuming subjective. Automated has gained interest to reduce clinician workload. Semantic involves labelling individual pixels in highlight infected regions. This article presents PulmonU-Net, innovative semantic model using PulmonNet modules as the base network areas chest X-rays. leverage global local characteristics create intricate feature maps. Incorporating leaky ReLU activation enables uninterrupted neuron functioning during learning. By adding encoder's deeper layers, addresses vanishing gradients improves dice similarity coefficient 94.25%. Real-time testing prediction visualization demonstrate PulmonU-Net's effectiveness automated infection

Язык: Английский

Процитировано

3
A Practical Study of Intelligent Image-Based Mobile Robot for Tracking Colored Objects DOI Open Access
Mofadal Alymani, Mohamed Esmail Karar, Hazem Ibrahim Shehata

и другие.

Computers, materials & continua/Computers, materials & continua (Print), Год журнала: 2024, Номер 80(2), С. 2181 - 2197

Опубликована: Янв. 1, 2024

Object tracking is one of the major tasks for mobile robots in many real-world applications. Also, artificial intelligence and automatic control techniques play an important role enhancing performance robot navigation. In contrast to previous simulation studies, this paper presents a new intelligent accomplishing multi-tasks by red-green-blue (RGB) colored objects real experimental field. Moreover, practical smart controller developed based on adaptive fuzzy logic custom proportional-integral-derivative (PID) schemes achieve accurate results, considering command delay tolerance errors. The design controllers implies some motion rules mimic knowledge experienced operators. Twelve scenarios three object combinations have been successfully tested evaluated using controlled image-based tracker. Classical PID failed handle study. proposed achieved best results with minimum average final error 13.8 cm reach targets, while our designed efficient saving both time traveling distance 6.6 s 14.3 cm, respectively. These promising demonstrate feasibility applying robotic system object-tracking environment reduce human workloads.

Язык: Английский

Процитировано

1
Hybrid transformer-CNN and LSTM model for lung disease segmentation and classification DOI Creative Commons

Syed Mohammed Shafi,

C. Sathiya Kumar

PeerJ Computer Science, Год журнала: 2024, Номер 10, С. e2444 - e2444

Опубликована: Дек. 13, 2024

According to the World Health Organization (WHO) report, lung disorders are third leading cause of mortality worldwide. Approximately three million individuals affected with various types annually. This issue alarms us take control measures related early diagnostics, accurate treatment procedures, etc. The precise identification through assessment medical images is crucial for pulmonary disease diagnosis. Also, it remains a formidable challenge due diverse and unpredictable nature pathological appearances shapes. Therefore, efficient segmentation classification model essential. By taking this initiative, novel hybrid LinkNet-Modified LSTM (L-MLSTM) proposed in research article. utilizes four essential fundamental steps its implementation. first step pre-processing, where input pre-processed using median filtering. Consequently, an improved Transformer-based convolutional neural network (CNN) (ITCNN) segment region process. After segmentation, features such as texture, shape, color, deep retrieved. Specifically, texture extracted modified Local Gradient Increasing Pattern (LGIP) Multi-texton analysis. Then, model, L-MLSTM model. work leverages two datasets COVID-19 normal pneumonia-CT dataset (Dataset 1) Chest CT scan 2). training evaluating providing comprehensive basis robust generalizable results. outperforms several existing models, including HDE-NN, DBN, LSTM, LINKNET, SVM, Bi-GRU, RNN, CNN, VGG19 + accuracies 89% 95% at learning percentages 70 90, respectively, 1 2. accuracy achieved by highlights capability better handle complexity variability images. approach enhances model's ability distinguish between different diseases reduces diagnostic errors compared methods.

Язык: Английский

Процитировано

1
VSMAS2HN: Verifiably Secure Mutual Authentication Scheme for Smart Healthcare Network DOI
Shivangi Batra, Bhawna Narwal, Amar Kumar Mohapatra

и другие.

Communications in computer and information science, Год журнала: 2023, Номер unknown, С. 150 - 160

Опубликована: Янв. 1, 2023

Язык: Английский

Процитировано

1
Segmenting and classifying lung diseases with M-Segnet and Hybrid Squeezenet-CNN architecture on CT images DOI Creative Commons

Syed Mohammed Shafi,

C. Sathiya Kumar

PLoS ONE, Год журнала: 2024, Номер 19(5), С. e0302507 - e0302507

Опубликована: Май 16, 2024

Diagnosing lung diseases accurately and promptly is essential for effectively managing this significant public health challenge on a global scale. This paper introduces new framework called Modified Segnet-based Lung Disease Segmentation Severity Classification (MSLDSSC). The MSLDSSC model comprises four phases: "preprocessing, segmentation, feature extraction, classification." Initially, the input image undergoes preprocessing using an improved Wiener filter technique. technique estimates power spectral density of noisy original images computes SNR assisted by PSNR to evaluate quality. Next, preprocessed identify separate RoI from background objects in image. We employ Segnet mechanism that utilizes proposed hard tanh-Softplus activation function effective Segmentation. Following Segmentation, features such as MLDN, entropy with MRELBP, shape features, deep are extracted. extraction phase, retrieved set into hybrid severity classification model. two classifiers: SDPA-Squeezenet DCNN. These classifiers train classify level diseases.

Язык: Английский

Процитировано

0