Gamma and X-Ray Technologies for Medical Research: Image Analysis and Disease Discovered

1. Introduction

The continuous pursuit of innovative diagnostic and medical research methods plays a crucial role in improving public health. Among the many available tools, technologies utilizing gamma and X-ray radiation form the foundation of modern medical imaging, enabling a minimally invasive yet detailed view inside the human body. Their application, extending from image analysis to the discovery of unknown aspects of diseases, is invaluable in medical research. Since Wilhelm Conrad Roentgen’s discovery of X-rays in 1895 [1], for which he was awarded the Nobel Prize in 1901, medical imaging has undergone immense evolution. X-ray radiation immediately found application in bone imaging and fracture detection. Decades after Röntgen’s discovery of X-rays, significant advancements in imaging technology led to the development of computed tomography (CT). This technique, for which its creators were recognized with a Nobel Prize in 1979, provides detailed three-dimensional images of tissues and organs. CT has revolutionized the diagnosis of many conditions, from cancers to internal injuries, offering unprecedented precision in localizing and assessing pathological changes [2].

In parallel with X-ray advancements, technologies based on gamma radiation have also evolved. Gamma radiation, which possesses higher energy than the X-rays used in medical diagnostics, is emitted by atomic nuclei during radioactive decay. Its primary application has been in nuclear medicine, where radiopharmaceuticals are introduced into a patient’s body. Gamma detectors, such as those used in positron emission tomography (PET) and single-photon emission computed tomography (SPECT), record the radiation emitted by these radioisotopes. These techniques allow for the assessment of metabolic and physiological functions of tissues, which is beyond the capabilities of traditional X-ray or CT scans. For instance, PET is extensively utilized in oncology for detecting metabolically active cancer cells, evaluating treatment effectiveness, and enabling the early detection of disease recurrence [3,4]. In neurology, both SPECT and PET are employed to study cerebral blood flow and receptor function, aiding in the diagnosis of neurodegenerative diseases like Alzheimer’s and Parkinson’s [5]. Both X-ray and gamma technologies necessitate advanced image analysis.

Contemporary diagnostic systems increasingly employ complex image processing algorithms, including artificial intelligence (AI) and machine learning (ML) techniques. The ever increasing volume of imaging data presents a significant challenge in current diagnostics. Consequently, artificial intelligence algorithms are becoming indispensable, as they can detect subtle patterns and anomalies that might be overlooked by the human eye. This capability significantly enhances diagnostic sensitivity and specificity. Experts often do not have enough time to thoroughly examine all medical diagnostic images.

For instance, deep learning algorithms are currently being tested for the automated detection of cancerous changes in chest X-rays and mammograms, as well as for analyzing PET images to assess disease progression [6,7]. The development of AI models, such as those based on convolutional neural networks (CNNs), has significantly improved the ability to segment and classify anatomical and pathological structures in medical images. This capability is crucial for the advancement of personalized medicine [8]. In the context of medical research, these technologies have opened new horizons. They allow scientists not only to better understand the pathophysiology of diseases at the molecular and cellular levels but also to develop novel therapeutic strategies. For example, studies using micro-CT on small laboratory animals carry out the real-time monitoring of disease progression, such as of osteoporosis or cancer, and assess experimental drug efficacy [9]. Similarly, PET and SPECT imaging in preclinical studies provide valuable information on the biodistribution and pharmacokinetics of investigational therapeutic agents, thereby accelerating the drug development process [10]. Recent achievements also include the integration of X-ray imaging with artificial intelligence techniques to identify disease biomarkers. This could lead to earlier and more precise diagnoses, for instance, in lung diseases [11]. As science continues to advance, we can anticipate further innovations in this field, which will contribute to improving the quality of life for patients worldwide [12]. The rapid analysis of the immense volume of data associated with the increasing number of available imaging systems is becoming an ever greater challenge.

2. An Overview of Published Articles

Significant progress in computer technologies over the last two decades has led to the integration of artificial intelligence (AI) and Monte Carlo (MC) methods into gamma and X-ray technologies for medical applications. Computer systems using AI and MC simulations have enabled the implementation of solutions, yielding numerous practical outcomes. In the context of radiography, AI plays a key role in improving diagnostic imaging. For instance, in lumbar spine radiography, oblique projections are frequently employed to assess spondylolysis and the morphology of facet joints. Crucially, the immediate determination of whether the oblique angle is appropriate for evaluating these conditions is essential. Moriya et al. [13] demonstrated the effectiveness of a convolutional neural network (CNN) in estimating the angle of oblique lumbar spine images immediately following radiography. This innovative application of AI helps address the critical need for prompt and accurate assessment of imaging angles in clinical practice.

Similarly, AI significantly enhances the work of radiologists in the evaluation of digital chest X-ray (CXR) images. This evaluation still relies primarily on visual confirmation by specialists, but the application of deep learning (DL) technologies will enable the immediate determination of the need for re-examination, consequently improving throughput. For such immediate analysis of CXR images, Usui et al. [14] developed three classification models (CLMs) for lung field defects, obstacle shadows, and the location of obstacle shadows, as well as a semantic segmentation model (SSM) for lung field areas, employing five-fold cross validation. AI also shows potential in detecting early signs of diseases such as breast cancer or cardiomegaly. The identification of calcifications in mammograms is critical for the early detection of breast cancer. Sakaida et al. [15] leveraged semi-supervised learning for this purpose, which combines a small dataset for supervised learning with deep learning. Their developed mammogram assessment method is expected to be an effective approach for automating the identification process.

Artificial intelligence also finds application in detecting cardiomegaly (i.e., heart enlargement) on CXRs. Two parameters, the cardiothoracic ratio (CTR) and the transverse cardiac diameter (TCD), are used to assess heart size on CXRs. Matusik et al. [16] investigated the performance and effectiveness of AI in cardiomegaly detection. Their evaluation suggests that AI can optimize the screening process for cardiomegaly on CXRs.

Mammography units must ensure optimal quality of generated images. Patients must often receive an elevated dose of ionizing radiation to achieve a mammogram of good quality. The challenge lies in minimizing this dose while maintaining optimal image quality. Szewczuk and Konefał [17], using Monte Carlo computer simulations, optimized the dose-to-image quality ratio for imaging with a selenium-based image detector, depending on breast composition and the voltage applied to the mammography unit’s X-ray tube for the used anode–filter systems.

3. Conclusions

The examples presented are merely the tip of the iceberg. The dynamic development of AI and MC, combined with their proven effectiveness in clinical practice, suggests immense potential for further innovation in medicine. These advancements will lead to even more personalized, more effective, and safer healthcare.