Medical Imaging Research Articles

An Efficient One-Dimensional Texture Representation Approach for Lung Disease Diagnosis

The remarkable increase in published medical imaging datasets for chest X-rays has significantly improved the performance of deep learning techniques to classify lung diseases efficiently. However, large datasets require special arrangements to make them suitable, accessible, and practically usable in remote clinics and emergency rooms. Additionally, it increases the computational time and image-processing complexity. This study investigates the efficiency of converting the 2D chest X-ray into one-dimensional texture representation data using descriptive statistics and local binary patterns, enabling the use of feed-forward neural networks to efficiently classify lung diseases within a short time and with cost effectiveness. This method bridges diagnostic gaps in healthcare services and improves patient outcomes in remote hospitals and emergency rooms. It also could reinforce the crucial role of technology in advancing healthcare. Utilizing the Guangzhou and PA datasets, our one-dimensional texture representation achieved 99% accuracy with a training time of 10.85 s and 0.19 s for testing. In the PA dataset, it achieved 96% accuracy with a training time of 38.14 s and a testing time of 0.17 s, outperforming EfficientNet, EfficientNet-V2-Small, and MobileNet-V3-Small. Therefore, this study suggests that the dimensional texture representation is fast and effective for lung disease classification.

Cervical Anatomical Characteristics in Women with Endometriosis: A Diagnostic Approach

Background: Endometriosis is a gynecologic disorder which causes dysmenorrhea and infertility. Early diagnosis of endometriosis can help prevent the necessity for invasive diagnostic procedures. Medical imaging has been widely utilized to diagnose various diseases without the need for invasive procedures. The purpose of this study was to investigate the cervical length in women with endometriosis. Methods: In this case-control study, the case group consisted of nulliparous women with endometriosis, while the control group comprised nulliparous women without endometriosis. A total of 42 individuals were included in each group. Cervical length was measured using transvaginal ultrasound from the external os to the internal os. The patients in the case group underwent laparoscopy to confirm the diagnosis. Pearson chi-square test and Fisher’s exact test were employed to compare categorical variables with a p<0.05 considered statistically significant. Results: In both groups, there were no notable variations in any of the demographic characteristics. However, the severity of dysmenorrhea was significantly different between the two groups (p=0.01). The average diameter of the mediolateral cervix (29.48±6.2 and 27.14±3.8) was statistically significant between the patient group and control group, respectively (p=0.04). The mediolateral width may have a positive predictive effect on the presence of endometriosis, while cervical length appears to have a protective effect against endometriosis. Conclusion: Demographic data do not predict endometriosis. This study suggests that mediolateral width in transvaginal sonography can serve as a minimally invasive diagnostic tool for endometriosis, showing correlation with endometriosis symptoms like dysmenorrhea and dyspareunia.

Enhanced breast cancer diagnosis through integration of computer vision with fusion based joint transfer learning using multi modality medical images

Enhanced breast cancer diagnosis through integration of computer vision with fusion based joint transfer learning using multi modality medical images

Privacy protection scheme for batch medical images based on double memristor cellular neural network

Privacy protection scheme for batch medical images based on double memristor cellular neural network

Quasi-Single-Crystal Tin Halide Perovskite Films with High Structural Integrity for Near-Infrared Imaging Array Enabling Hidden Object Recognition.

Near-infrared (NIR) image sensors based on solution-processed thin film are gaining prominence in various applications, including security detection, remote sensing, medical imaging, and environmental monitoring, owing to superior penetration capabilities of NIR light. However, the reported perovskite image sensors suffer from limited resolution and performance due to poor structural integrity, bioincompatibility, and constrained response wavelength range from lead-based perovskite materials employed. In this study, we present non-toxic quasi-single-crystal (QSC) CH(NH2)2SnI3 (FASnI3) perovskite films, prepared via a simple spin-coating technique, that demonstrate high structural integrity and effective NIR response. Through a series of in situ characterizations, we reveal that the orderly growth mode of QSC-FASnI3 perovskite films enables a more oriented crystal growth with reduced trap and grain boundaries. Consequently, self-powered NIR photodetector based on QSC-FASnI3 films exhibited large detectivity over 1013 Jones in the NIR range (780-890 nm). Ultimately, due to the high structural integrity, the high-resolution 64×64 (4096) pixels NIR imaging array, exhibiting reduced photo and dark response non-uniformity value, was fabricated on the thin-film transistors (TFT) backplanes, enabling ultraweak NIR light (63 nW cm-2) real-time imaging, fingerprint imaging, and hidden object recognition. This work pioneers the application of lead-free perovskite for high-resolution NIR imaging arrays.

Quasi‐Single‐Crystal Tin Halide Perovskite Films with High Structural Integrity for Near‐Infrared Imaging Array Enabling Hidden Object Recognition

AbstractNear‐infrared (NIR) image sensors based on solution‐processed thin film are gaining prominence in various applications, including security detection, remote sensing, medical imaging, and environmental monitoring, owing to superior penetration capabilities of NIR light. However, the reported perovskite image sensors suffer from limited resolution and performance due to poor structural integrity, bioincompatibility, and constrained response wavelength range from lead‐based perovskite materials employed. In this study, we present non‐toxic quasi‐single‐crystal (QSC) CH(NH2)2SnI3 (FASnI3) perovskite films, prepared via a simple spin‐coating technique, that demonstrate high structural integrity and effective NIR response. Through a series of in situ characterizations, we reveal that the orderly growth mode of QSC‐FASnI3 perovskite films enables a more oriented crystal growth with reduced trap and grain boundaries. Consequently, self‐powered NIR photodetector based on QSC‐FASnI3 films exhibited large detectivity over 1013 Jones in the NIR range (780–890 nm). Ultimately, due to the high structural integrity, the high‐resolution 64×64 (4096) pixels NIR imaging array, exhibiting reduced photo and dark response non‐uniformity value, was fabricated on the thin‐film transistors (TFT) backplanes, enabling ultraweak NIR light (63 nW cm−2) real‐time imaging, fingerprint imaging, and hidden object recognition. This work pioneers the application of lead‐free perovskite for high‐resolution NIR imaging arrays.

A Computational Protocol for the Knowledge-Based Assessment and Capture of Pathologies.

We propose that one of the main hurdles in delivering comprehensively informed care results from the challenges surrounding the extraction, representation, and retention of prior clinical experience and basic medical knowledge, as well as its translation into time- and context-informed actionable interventions. While emerging applications in artificial intelligence-based techniques, for example, large language models, offer impressive pattern association capabilities, they often fall short in producing human-readable explanations crucial to their integration into clinical care. Moreover, they require large well-defined and well-integrated data sets that typically conflict with the availability of such data in all but a few areas of medicine, for example, medical imaging and neuroimaging, noninvasive monitoring of bio-electrical activity, etc. In this chapter, we argue that approximate reasoning rooted in the knowledge that is explainable to the human clinician may offer attractive avenues for the introduction of such knowledge in a systematic way that supports formal retention, sharing, and reuse of new clinical and basic medical experience. We outline a conceptual protocol that targets the use of sparse and disparate data of different types and from different sources, seamlessly drawing on our collective experience and that of others. We illustrate the utility of such an integrative approach by applying the latter to the assessment and reconciliation of data from different experimental models, human and animal, in the example use case of a complex health condition.

Kruskal Szekeres generative adversarial network augmented deep autoencoder for colorectal cancer detection.

Cancer involves abnormal cell growth, with types like intestinal and oesophageal cancer often diagnosed in advanced stages, making them hard to cure. Symptoms are like burning sensations in the stomach and swallowing difficulties are specified as colorectal cancer. Deep learning significantly impacts the medical image processing and diagnosis, offering potential improvements in accuracy and efficiency. The Kruskal Szekeres Generative Adversarial Network Augmented Deep Autoencoder (KSGANA-DA) is introduced for early colorectal cancer detection and it comprises two stages; Initial stage, data augmentation uses Affine Transform via Random Horizontal Rotation and Geometric Transform via Kruskal-Szekeres that coordinates to improve the training dataset diversity, boosting detection performance. The second stage, a Deep Autoencoder Anatomical Landmark-based Image Segmentation preserves edge pixel spatial locations, improving precision and recall for early boundary detection. Experiments validate KSGANA-DA performance and different existing methods are implemented into Python. The results of KSGANA-DA are to provide higher precision by 41%, recall by 7%, and lesser training time by 46% than compared to conventional methods.

Advancing clinical MRI exams with artificial intelligence: Japan's contributions and future prospects.

In this narrative review, we review the applications of artificial intelligence (AI) into clinical magnetic resonance imaging (MRI) exams, with a particular focus on Japan's contributions to this field. In the first part of the review, we introduce the various applications of AI in optimizing different aspects of the MRI process, including scan protocols, patient preparation, image acquisition, image reconstruction, and postprocessing techniques. Additionally, we examine AI's growing influence in clinical decision-making, particularly in areas such as segmentation, radiation therapy planning, and reporting assistance. By emphasizing studies conducted in Japan, we highlight the nation's contributions to the advancement of AI in MRI. In the latter part of the review, we highlight the characteristics that make Japan a unique environment for the development and implementation of AI in MRI examinations. Japan's healthcare landscape is distinguished by several key factors that collectively create a fertile ground for AI research and development. Notably, Japan boasts one of the highest densities of MRI scanners per capita globally, ensuring widespread access to the exam. Japan's national health insurance system plays a pivotal role by providing MRI scans to all citizens irrespective of socioeconomic status, which facilitates the collection of inclusive and unbiased imaging data across a diverse population. Japan's extensive health screening programs, coupled with collaborative research initiatives like the Japan Medical Imaging Database (J-MID), enable the aggregation and sharing of large, high-quality datasets. With its technological expertise and healthcare infrastructure, Japan is well-positioned to make meaningful contributions to the MRI-AI domain. The collaborative efforts of researchers, clinicians, and technology experts, including those in Japan, will continue to advance the future of AI in clinical MRI, potentially leading to improvements in patient care and healthcare efficiency.

Using Compressed JPEG and JPEG2000 Medical Images in Deep Learning: A Review

Machine Learning (ML), particularly Deep Learning (DL), has become increasingly integral to medical imaging, significantly enhancing diagnostic processes and treatment planning. By leveraging extensive datasets and advanced algorithms, ML models can analyze medical images with exceptional precision. However, their effectiveness depends on large datasets, which require extended training times for accurate predictions. With the rapid increase in data volume due to advancements in medical imaging technology, managing the data has become increasingly challenging. Consequently, irreversible compression of medical images has become essential for efficiently handling the substantial volume of data. Extensive research has established recommended compression ratios tailored to specific anatomies and imaging modalities, and these guidelines have been widely endorsed by government bodies and professional organizations globally. This work investigates the effects of irreversible compression on DL models by reviewing the relevant literature. It is crucial to understand how DL models respond to image compression degradations, particularly those introduced by JPEG and JPEG2000—both of which are the only permissible irreversible compression techniques in the most commonly used medical image format—the Digital Imaging and Communications in Medicine (DICOM) standard. This study provides insights into how DL models react to such degradations, focusing on the loss of high-frequency content and its implications for diagnostic interpretation. The findings suggest that while existing studies offer valuable insights, future research should systematically explore varying compression levels based on modality and anatomy, and consider developing strategies for integrating compressed images into DL model training for medical image analysis.

Cutting-Edge Deep Learning Strategies for Precise Ovarian Disease Detection: A Comprehensive Review

Accurate and timely diagnosis of ovary-related diseases is of paramount importance in the realm of medical imaging. Utilizing deep learning technology has emerged as a viable method to improve the precision and effectiveness of medical image segmentation, which directly impacts the detection of ovarian diseases. Ovarian disease detection through medical image segmentation is critical for early diagnosis and effective treatment. This study evaluates cutting-edge deep learning strategies using the MMOTU ovarian tumour ultrasound dataset, which includes OTU 2D and OTU CEUS subsets. We implemented various models, including U-Net, its variants with ResNet and DenseNet backbones, CR-Unet, and Ocys-Net, along with ensemble learning and transfer learning techniques. Results show that while the baseline U-Net is effective, advanced models, particularly CR-Unet and DenseNet, significantly improve segmentation accuracy. The best performance was achieved using an ensemble model and a fine-tuned pre-trained network, highlighting the potential of these approaches in enhancing the precision and reliability of ovarian disease detection.

Enhancing Text-to-Image Generation: Integrating CLIP and Diffusion Models for Improved Visual Accuracy and Semantic Consistency

Abstract. Text-to-Image (T2I) generation focuses on producing images that precisely match given textual descriptions by combining techniques from computer vision and natural language processing (NLP). Existing studies have shown an innovative approach to enhance T2I generation by integrating Contrastive Language-Image Pretraining (CLIP) embeddings with a Diffusion Model (DM). The method involves initially extracting rich and meaningful text embeddings using CLIP, which are then transformed into corresponding images. These images are progressively refined through an iterative denoising process enabled by diffusion models. Comprehensive experiments conducted on the MS-COCO dataset validate the proposed method, demonstrating significant improvements in image fidelity and the alignment between text and images. When compared to traditional models like Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), which often struggle with maintaining both visual quality and semantic accuracy, this hybrid model shows superior performance. Future research could explore optimizing hybrid models further and applying T2I technology to specialized fields, such as medical imaging and scientific visualization, expanding its potential use cases.

BCDNet: A Deep Learning Model with Improved Convolutional Neural Network for Efficient Detection of Bone Cancer Using Histology Images

Among the several types of cancer, bone cancer is the most lethal prevailing in the world. Its prevention is better than cure. Besides early detection of bone cancer has potential to have medical intervention to prevent spread of malignant cells and help patients to recover from the disease. Many medical imaging modalities such as histology, histopathology, radiology, X-rays, MRIs, CT scans, phototherapy, PET and ultrasounds are being used in bone cancer detection research. However, hematoxylin and eosin stained histology images are found crucial for early diagnosis of bone cancer. Existing Convolutional Neural Network (CNN) based deep learning techniques are found suitable for medical image analytics. However, the models are prone to mediocre performance unless configured properly with empirical study. Within this article, we suggested a framework centered on deep learning for automatic bone cancer detection. We also proposed a CNN variant known as Bone Cancer Detection Network (BCDNet) which is configured and optimized for detection of a common kind of bone cancer named Osteosarcoma. An algorithm known as Learning based Osteosarcoma Detection (LbOD). It exploits BCDNet model for both binomial and multi-class classification. Osteosarcoma-Tumor-Assessment is the histology dataset used for our empirical study. Our the outcomes of the trial showed that BCDNet outperforms baseline models with 96.29% accuracy in binary classification and 94.69% accuracy in multi-class classification.

Medical Image Segmentation

Medical image segmentation is a critical component in the development of computer-aided diagnosis and treatment planning systems. This paper provides a comprehensive survey of recent advances in segmentation techniques applied to various imaging modalities, including Magnetic Resonance Imaging (MRI). Traditional methods such as thresholding, region-growing, and active contours are reviewed alongside contemporary machine learning-based approaches, particularly deep learning models. The survey emphasizes the growing dominance of convolutional neural networks (CNNs) and their variants, including U-Net and Fully Convolutional Networks (FCNs), which have shown remarkable success in handling complex medical imaging challenges. Additionally, the paper discusses hybrid methods that combine classical techniques with artificial intelligence to improve accuracy and robustness in segmentation tasks. Key challenges such as class imbalance, boundary delineation, and computational efficiency are also highlighted. Future directions, including the integration of multi-modal data and advancements in self-supervised learning, are explored as potential solutions to overcome current limitations in medical image segmentation.

Multiwavelength photobiomodulation in atrophic age-related macular degeneration.

To evaluate photobiomodulation (PBM) safety and efficacy in patient with atrophic AMD and to explore tissue effects using Spectrally Resolved Autofluorescence (SrAF) to identify a potential biomarker indicative of mitochondrial activity, the primary PBM target. This retrospective, non-comparative case series involved six eyes of five patients with atrophic AMD, conducted at the Medical Retina and Imaging Unit, IRCCS San Raffaele Hospital in Milan, Italy. At baseline and follow-ups (after one and three months) a complete ophthalmological assessment and multimodal imaging, including spectrally resolved autofluorescence (SrAF), were performed. PBM treatment (Valeda Light Delivery System - LumiThera) was administered in nine sessions over three weeks. The SrAF images were analyzed to evaluate the effect of the treatment. Six eyes of five patients with atrophic AMD received PBM at baseline. Best corrected visual acuity (BCVA), mean central macular thickness (CMT), mean retinal sensitivity on microperimetry and the atrophic area have remained stable. No eyes developed foveal atrophy after the treatment. The image analysis in SrAF showed an increase in the highlighted area of 4.89% after one month and 17.11% after three months. No adverse systemic or local side effects were reported. PBM is a safe treatment for the human retina. The evaluation of the retinal area within the green light spectrum (575 nm) in SrAF is a potential indirect biomarker of mitochondrial activity and could be used to assess treatment efficacy in clinical studies.

An intelligent magnetic resonance imagining-based multistage Alzheimer's disease classification using swish-convolutional neural networks.

Alzheimer's disease (AD) refers to a neurological disorder that causes damage to brain cells and results in decreasing cognitive abilities and memory. In brain scans, this degeneration can be seen in different ways. The disease can be classified into four stages: Non-demented (ND), moderate demented (MoD), mild demented (MiD), and very mild demented (VMD). To prepare the raw dataset for analysis, the collected magnetic resonance imaging (MRI) images are subjected to several pre-processing techniques in order to improve the performance accuracy of the proposed model. Medical images generally have poor contrast and get affected by noise, which ends up with inaccurate diagnosis. For the different phases of AD to be detected, a clear image is necessary. To address this issue, the influence of the artefacts must be reduced, enhance the contrast, and reduce the loss of information. A novel framework for image enhancement is suggested to increase the accuracy in the detection and identification of AD. In this study, the raw MRI dataset from the Alzheimer's disease neuroimaging initiative (ADNI) database is subjected to skull stripping, contrast enhancement, and image filtering followed by data augmentation to balance the dataset with four types of Alzheimer's classes. The pre-processed data are subjected to five different pre-trained models such as AlexNet, ResNet, VGG 16, EfficientNet, and Inceptionv3 achieving a testing accuracy rate of 91.2%, 88.21%, 92.34%, 93.45%, and 85.12%, respectively. These pre-trained models are compared with the proposed CNN (convolutional neural network) model designed with Adam optimizer and Flatten Swish activation function which reaches the highest accuracy of 96.5% with a learning rate of 0.000001. The five pre-trained CNNmodels along with the proposed swish-based AD-CNN were tested using various performance metrics to evaluate the model efficiency in classifying and identifying the AD classes. From the result analysis, it is evident that the proposed AD-CNN model outperforms all the other models.

CNN-Based Cancer Image Diagnosis: Current Progress and Future Directions

Abstract. With the development of deep learning technology, CNN (Convolutional Neural Network) models have shown great value in medical image analysis, especially in diagnosing early lung cancer, breast cancer, and brain tumors. In this study, we recall and organize the application and progress of CNN models in the field of cancer and tumor diagnosis in the past five years to provide a theoretical basis and reference for related researchers. This article introduces the principles of different CNN cancer diagnostic models and compares and analyzes their results, and ultimately finds that these models have significant advantages in improving the accuracy and efficiency of cancer diagnosis, but at the same time, there are also problems such as too much reliance on large datasets, high model complexity, and poor generalization ability. In the future, we can consider optimizing the performance of CNN network models, enhancing the generalization ability of the models, and developing data enhancement techniques on this basis, so that CNN models can be better applied in the field of cancer diagnosis.

Depth-resolved imaging through scattering media using time-gated light field tomography.

We present a novel, to the best of our knowledge, approach to overcome the limitations imposed by scattering media using time-gated light field tomography. By integrating the time-gating technique with light field imaging, we demonstrate the ability to capture and reconstruct images with different depths through highly scattering environments. Our method exploits the temporal characteristics of light propagation to selectively isolate ballistic photons, enabling enhanced depth resolution and improved imaging quality. Through comprehensive experimental validation and analysis, we showcase the effectiveness of our technique in resolving depth information with high fidelity, even in the presence of significant scattering. The resultant system can simultaneously acquire multi-angled projections of the object without requiring prior knowledge of the media or the target. This advancement holds promise for a wide range of applications, including non-invasive medical imaging, environmental monitoring, and industrial inspection, where imaging through scattering media is critical for an accurate and reliable analysis.

Imagine Denoising Methods Based on GANs

Abstract. With the continuous development of technology, peoples demand for image clarity is also constantly increasing. Whether in the fields of medical imaging, aerospace, or people's daily lives, noise in images seriously affects their clarity. Therefore, how to remove noise on the premise of preserving image details has now become a hot research topic. In the domain of image denoising, both early spatial domain filtering methods and recently proposed convolutional neural network models have certain limitations. Compared to other denoising methods, GANs can better remove noise from images and improve image quality. Therefore, this article summarizes and organizes image denoising methods based on GAN models. This article explains four GAN based image denoising methods, namely GAN, WGAN, DNGAN, and GCBD, from the perspectives of framework structure, advantages and disadvantages, and application fields. At the same time, this article also analyzes the development trend of GAN application in image denoising and makes prospects.

A compact X-ray source via fast microparticle streams.

The spatiotemporal resolution of diagnostic X-ray images is limited by the erosion and rupture of conventional stationary and rotating anodes of X-ray tubes from extreme density of input power and thermal cycling of the anode material. Conversely, detector technology has developed rapidly. Finer detector pixels demand improved output from brilliant keV-type X-ray sources with smaller X-ray focal spots than today and would be available to improve the efficacy of medical imaging. In addition, novel cancer therapy demands for greatly improved output from X-ray sources. However, since its advent in 1929, the technology of high-output compact X-ray tubes has relied upon focused electrons hitting a spinning rigid rotating anode; a technology that, despite of substantial investment in material technology, has become the primary bottleneck of further improvement. In the current study, an alternative target concept employing a stream of fast discrete metallic microparticles that intersect with the electron beam is explored by simulations that cover the most critical uncertainties. The concept is expected to have far-reaching impact in diagnostic imaging, radiation cancer therapy and non-destructive testing. We outline technical implementations that may become the basis of future X-ray source developments based on the suggested paradigm shift.