Accuracy Of Medical Imaging Research Articles

Application Exploration of the Photon Chip Technology in Real time Processing of Biomedical Images

In the biomedical field, photonic chip technology is gradually becoming the key to improving the speed and accuracy of medical imaging. The development of this technology has given us new hope in early disease diagnosis and cell monitoring. Recent research has not only sorted out the hotspots and future trends of biomedical photonics, but also paid special attention to the application of photon technology in these fields. Specifically, photonic chips have demonstrated their unique capabilities in multi-color two-photon excitation fluorescence microscopy imaging, two-photon fluorescence lifetime imaging, and two-photon fiber endoscopic imaging. These applications not only improve the clarity of imaging, but also accelerate processing speed, which is crucial for rapid diagnosis and treatment. At the end of the article, the author summarized the contributions of photonic chip technology in the field of biomedical imaging and provided prospects for future development directions.

Explainable lung cancer classification with ensemble transfer learning of VGG16, Resnet50 and InceptionV3 using grad-cam

Medical imaging stands as a critical component in diagnosing various diseases, where traditional methods often rely on manual interpretation and conventional machine learning techniques. These approaches, while effective, come with inherent limitations such as subjectivity in interpretation and constraints in handling complex image features. This research paper proposes an integrated deep learning approach utilizing pre-trained models—VGG16, ResNet50, and InceptionV3—combined within a unified framework to improve diagnostic accuracy in medical imaging. The method focuses on lung cancer detection using images resized and converted to a uniform format to optimize performance and ensure consistency across datasets. Our proposed model leverages the strengths of each pre-trained network, achieving a high degree of feature extraction and robustness by freezing the early convolutional layers and fine-tuning the deeper layers. Additionally, techniques like SMOTE and Gaussian Blur are applied to address class imbalance, enhancing model training on underrepresented classes. The model’s performance was validated on the IQ-OTH/NCCD lung cancer dataset, which was collected from the Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases over a period of three months in fall 2019. The proposed model achieved an accuracy of 98.18%, with precision and recall rates notably high across all classes. This improvement highlights the potential of integrated deep learning systems in medical diagnostics, providing a more accurate, reliable, and efficient means of disease detection.

Exploring the Impact of Noise and Image Quality on Deep Learning Performance in DXA Images.

Segmentation of the femur in Dual-Energy X-ray (DXA) images poses challenges due to reduced contrast, noise, bone shape variations, and inconsistent X-ray beam penetration. In this study, we investigate the relationship between noise and certain deep learning (DL) techniques for semantic segmentation of the femur to enhance segmentation and bone mineral density (BMD) accuracy by incorporating noise reduction methods into DL models. Convolutional neural network (CNN)-based models were employed to segment femurs in DXA images and evaluate the effects of noise reduction filters on segmentation accuracy and their effect on BMD calculation. Various noise reduction techniques were integrated into DL-based models to enhance image quality before training. We assessed the performance of the fully convolutional neural network (FCNN) in comparison to noise reduction algorithms and manual segmentation methods. Our study demonstrated that the FCNN outperformed noise reduction algorithms in enhancing segmentation accuracy and enabling precise calculation of BMD. The FCNN-based segmentation approach achieved a segmentation accuracy of 98.84% and a correlation coefficient of 0.9928 for BMD measurements, indicating its effectiveness in the clinical diagnosis of osteoporosis. In conclusion, integrating noise reduction techniques into DL-based models significantly improves femur segmentation accuracy in DXA images. The FCNN model, in particular, shows promising results in enhancing BMD calculation and clinical diagnosis of osteoporosis. These findings highlight the potential of DL techniques in addressing segmentation challenges and improving diagnostic accuracy in medical imaging.

Collaborative Enhancement of Consistency and Accuracy in US Diagnosis of Thyroid Nodules Using Large Language Models.

Background Large language models (LLMs) hold substantial promise for medical imaging interpretation. However, there is a lack of studies on their feasibility in handling reasoning questions associated with medical diagnosis. Purpose To investigate the viability of leveraging three publicly available LLMs to enhance consistency and diagnostic accuracy in medical imaging based on standardized reporting, with pathology as the reference standard. Materials and Methods US images of thyroid nodules with pathologic results were retrospectively collected from a tertiary referral hospital between July 2022 and December 2022 and used to evaluate malignancy diagnoses generated by three LLMs-OpenAI's ChatGPT 3.5, ChatGPT 4.0, and Google's Bard. Inter- and intra-LLM agreement of diagnosis were evaluated. Then, diagnostic performance, including accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC), was evaluated and compared for the LLMs and three interactive approaches: human reader combined with LLMs, image-to-text model combined with LLMs, and an end-to-end convolutional neural network model. Results A total of 1161 US images of thyroid nodules (498 benign, 663 malignant) from 725 patients (mean age, 42.2 years ± 14.1 [SD]; 516 women) were evaluated. ChatGPT 4.0 and Bard displayed substantial to almost perfect intra-LLM agreement (κ range, 0.65-0.86 [95% CI: 0.64, 0.86]), while ChatGPT 3.5 showed fair to substantial agreement (κ range, 0.36-0.68 [95% CI: 0.36, 0.68]). ChatGPT 4.0 had an accuracy of 78%-86% (95% CI: 76%, 88%) and sensitivity of 86%-95% (95% CI: 83%, 96%), compared with 74%-86% (95% CI: 71%, 88%) and 74%-91% (95% CI: 71%, 93%), respectively, for Bard. Moreover, with ChatGPT 4.0, the image-to-text-LLM strategy exhibited an AUC (0.83 [95% CI: 0.80, 0.85]) and accuracy (84% [95% CI: 82%, 86%]) comparable to those of the human-LLM interaction strategy with two senior readers and one junior reader and exceeding those of the human-LLM interaction strategy with one junior reader. Conclusion LLMs, particularly integrated with image-to-text approaches, show potential in enhancing diagnostic medical imaging. ChatGPT 4.0 was optimal for consistency and diagnostic accuracy when compared with Bard and ChatGPT 3.5. © RSNA, 2024 Supplemental material is available for this article.

Adaptive-sized residual fusion network-based segmentation of biomedical images

Medical imaging is crucial in examinations and for providing appropriate patient care. Medical images are used to diagnose and identify specific conditions. However, the diagnostic procedures are confounded by the complexity of these images, which have several overlapping characteristics. Various techniques have enhanced biomedical image segmentation, and strategies to improve the contrast and accuracy of medical images have been presented in the literature. This article presents a recurrent grasshopper herd-based optimized histogram-based contrast enhancement technique, in which sharp-edge morphological characteristics can be enhanced using inclusive Laplacian fractional order Savitzky–Golay differentiation. An adaptive-sized residual fusion network approach is used to isolate the target tissue in the biomedical image. The effectiveness of the proposed system is verified using different medical data sets and a high level of accuracy is achieved. The proposed method outperforms other state-of-the-art methodologies in terms of the scale, shape and morphological attributes of the medical image.

Description of the Field of Radiology in Medical Imaging

In medicine, the role of medical images in the field of radiology is very important. Accuracy of medical images and high quality of images are very important to the attending physician in making an accurate diagnosis of the patient. Processing, storage, transformation of medical images, conversion from analog to digital form.

Pneumonia Detection System Using Deep Learning

Artificial intelligence and machine learning are increasingly being applied in medicine, particularly in biomedical imaging and diagnostic procedures. Machine learning algorithms are being used to process chest X-ray images, enhancing consistency and accuracy in reporting. The research focuses on using deep learning algorithms based on convolutional neural networks to build a processing model for detecting pneumonia-related changes in chest X-rays and classifying them into two groups based on detection results. This approach aims to improve decision-making and accuracy in medical imaging.

Performance of ultra‐wide band DCBLNA with suspended strip line radiator for human breast cancer diagnosis medical imaging application

This study presents the performance of differential cascode balun low noise amplifier (DCBLNA) with ultra-wideband (UWB) for human breast cancer diagnosis. The proposed DCBLNA design-I with bulky spiral inductors achieves insufficient bandwidth with large power consumption of 10.8 mW. To attain the proper UWB band of operation, suspended strip line (SSLIN) radiators have employed in the proposed design-I. The performance of SSLIN is evaluated in terms of line capacitance and characteristic impedance by optimising its width. It is observed that best 50 Ωn-II. DCBLNA design-II using SSLIN have achieving a desired band of operation ranging from 1.5 to 15.7 GHz and best NF of 0.5 dB. The gain and phase imperfections are simulated to characterise balun networks. The smallest gain imperfection achieved is 0.1 dB at 10 GHz while the simulated phase imperfection turns out to be sufficiently good with 2.35° at 8 GHz. The proposed DCBLNA design-II is implemented and fabricated using RFCMOS 45 nm Taiwan Semiconductor Manufacturing Company (TSMC) process under commercial conditions. The highest figure of merit comes out to be 3.2 that ensures good accuracy of medical imaging for breast cancer diagnosis.

A DE-ANN Inspired Skin Cancer Detection Approach Using Fuzzy C-Means Clustering

As per recent developments in medical science, the skin cancer is considered as one of the common type disease in human body. Although the presence of melanoma is viewed as a form of cancer, it is challenging to predict it. If melanoma or other skin diseases are identified in the early stages, prognosis can then be successfully achieved to cure them. For this, medical imaging science plays an essential role in detecting such types of skin lesions quickly and accurately. The application of our approaches is to improve skin cancer detection accuracy in medical imaging and further, can be automated using electronic devices such as mobile phones etc. In the proposed paper, an improved strategy to detect three type of skin cancers in early stages are suggested. The considered input is a skin lesion image which by using the proposed method, the system would classify it into cancerous or non-cancerous type of skin. The image segmentation is implemented using fuzzy C-means clustering to separate homogeneous image regions. The preprocessing is done using different filters to enhance the image attributes while the other features are assessed by implementing rgb color-space, Local Binary Pattern (LBP) and GLCM methods altogether. Further, for classification, artificial neural network (ANN) is trained using differential evolution (DE) algorithm. Various features are accurately estimated to achieve better results using skin cancer image datasets namely HAM10000 and PH2. The novelty of the work suggests that DE-ANN is best compared among other traditional classifiers in terms of detection accuracy as discussed in result section of this paper. The simulated result shows that the proposed technique effectually detects skin cancer and produces an accuracy of 97.4%. The results are highly accurate compare to other traditional approaches in the same domain.

Theoretical assessment of electro-thermal imaging: A new technique for medical diagnosis

Breast cancer is one of the most crucial cancer types. To improve the diagnosis performance, a hybrid system is proposed through simultaneous utilization of thermal and electrical impedance imaging methods. The innovation of the approach relies on the frequency dependence of the tissue’s electrical impedance which facilitates the acquisition of multiple thermal images with currents at different frequencies injected to the region of the body under inspection. The applied current and the resulting heating at the body surface are distributed based on the frequency dependent conductivity distribution. The electrical currents increase the thermal contrast on the body surface depending on the electrical properties of the tissues at the operation frequency. The technique also provides frequency dependent conductivity distribution data through thermal imaging which can be used as a basis for the detection of the breast carcinoma. Based on our findings, the contrast resolution between the healthy and cancerous tissue is increased, improving the depth-dependent imaging performance from 3mm to 9mm for a 1.5mm tumor. The sensitivity of the technique can be further increased by an infrared camera with dual band imaging capability. Consequently, the proposed approach has a potential to improve the sensitivity and accuracy of medical imaging over the standard thermography.

Investigation of Distance, Machine and Kernel Learning Similarity Methods for Visual Search in Content based Image Retrieval

Background/Objectives: The major objective of this work is to increase the retrieval accuracy of medical images by measuring the visual similarity of Content-Based Image Retrieval (CBIR) system. Methods/Statistical Analysis: This paper presents an On-line Multiple Kernel Similarity (OMKS) learning framework for performing classification based on kernel-based proximity functions. In accordance with many existing methods some other related issues are also discussed and retrieval performance evaluation of the existing and proposed OMKS learning strategy is also discussed. Findings: Several number of the distance based learning algorithm has been proposed in recent works. It is mainly aimed for the sake of measuring the visual similarity corresponding to the images. This paper involves in providing a comprehensive review of the technical achievements of distance learning, machine learning along with kernel learning methods for conducting visual similarity search. The methods cited have limitation in their capability of the similar measurement with complicated patterns in most practical applications. Similarity of multimodal data identified through the multiple resources 28 , cannot be handled. Application/Improvements: Evaluation of the technique proposed for CBIR is performed on a huge amount of image data sets where motivating results indicate that OMKS performs better than the state-of the- art techniques significantly. At last, based on OMKS technology and the rise of requisitions from practical-world applications, and idealistic future research directions have been identified as suggestions 29 .

Accuracy of internet images of glenoid labral injuries

The internet is a convenient source of health information used widely by patients and doctors. Previous studies have found that the written information provided was often inaccurate. There is no literature regarding the accuracy of medical images on the internet. The aim of this study was to assess the accuracy of internet images of injuries to the glenoid labrum following shoulder dislocation. The Google and Bing search engines were used to find images of Bankart, Perthes and anterior labroligamentous periosteal sleeve avulsion (ALPSA) lesions. Three independent reviewers assessed the accuracy of image labelling. Of images labelled 'Bankart lesion', 30% (9/30) were incorrect while 'Perthes lesion' images were incorrect in 15% of cases (9/60) and 4% of 'ALPSA lesion' images were incorrect (2/46). There was good interobserver reliability (kappa = 0.81). Labelling accuracy was better on educational sites than on commercial sites (6% vs 25% inaccurate, p=0.0013). Caution is recommended when interpreting non-peer reviewed images on the internet.

Patient confidentiality issues and accuracy of medical images used in publications

Patient confidentiality issues and accuracy of medical images used in publications

Reply to ‘Patient confidentiality issues and accuracy of medical images used in publications’

Reply to ‘Patient confidentiality issues and accuracy of medical images used in publications’

Assessing diagnostic accuracy in veterinary imaging.

With continued specialization in diagnostic radiology and imaging technology, it increasingly becomes important for diagnostic radiologists to focus on the science of diagnosing disease. Imaging modalities provide a means to examine patients; however, it is the detection of imaging signs (Roentgen signs) that constitutes imaging tests. All tests are subject to error, and the degree of error cannot be determined by clinical experience alone. Therefore, measurements of diagnostic accuracy are important. Accuracy estimates are greatly influenced by selection of the decision criterion and sample population. This article reviews some of the principles for assessing diagnostic accuracy in medical imaging, which is helpful for deciding what tests to perform and how to interpret results once a test has been performed.

The EASI project--improving the effectiveness and quality of image-guided surgery.

In recent years, advances in computer technology and a significant increase in the accuracy of medical imaging have made it possible to develop systems that can assist the clinician in diagnosis, planning, and treatment. This paper deals with an area that is generally referred to as computer-assisted surgery, image-directed surgery, or image-guided surgery. We report the research, development, and clinical validation performed since January 1996 in the European Applications in Surgical Interventions (EASI) project, which is funded by the European Commission in their "4th Framework Telematics Applications for Health" program. The goal of this project is the improvement of the effectiveness and quality of image-guided neurosurgery of the brain and image-guided vascular surgery of abdominal aortic aneurysms, while at the same time reducing patient risks and overall cost. We have developed advanced prototype systems for preoperative surgical planning and intraoperative surgical navigation, and we have extensively clinically validated these systems. The prototype systems and the clinical validation results are described in this paper.