Magnetic resonance spectroscopy is a noninvasive method that enables analysis of the metabolic profile of liver tissue and the identification of biochemical indicators of malignant transformation. In this review we summarize current research data on the use of magnetic resonance spectroscopy in the differential diagnosis of benign and malignant focal liver lesions and to assess the diagnostic value of metabolic indices. A search and selection of publications were performed in the PubMed, Web of Science, and Google Scholar databases using the following keywords: "magnetic resonance spectroscopy," "MRS," "liver lesion," "MRI," "choline/lipid," and "MRS metabolites." The analysis included articles published from inception of the said databases up to December 2024 that focused on the application of magnetic resonance spectroscopy for evaluating the metabolic profile of liver tumors. Numerous studies indicate that magnetic resonance spectroscopy allows for the detection of characteristic changes in the levels of choline, lipids, and other metabolites reflecting membrane and energy exchange processes. These parameters offer additional noninvasive criteria for differentiating benign and malignant liver lesions. Although routine clinical use of magnetic resonance spectroscopy remains limited, the method demonstrates high potential for the early detection of tumors and the assessment of their metabolic characteristics. Further research may contribute to the technique standardization and a deeper understanding of the biochemical mechanisms underlying tumor growth, which is particularly important in complex and diagnostically challenging cases.
BACKGROUND: Bladder cancer is the most common malignant neoplasm of the urinary system. The mean age at diagnosis is 73 years, which suggests high risks of comorbidity and clinical risks during invasive diagnostic procedures. Tumor invasion into musculature is a key factor governing treatment choice that necessitates a histological examination. The overall quality of histological examinations relies heavily on biopsy sample adequacy. In this regard, the development and implementation of novel biomarkers based on advanced imaging techniques remain highly relevant, as they may improve the accuracy of T-stage assessment and help predict the clinical course of bladder cancer. From this perspective, the potential application of magnetic resonance imaging texture analysis is being actively discussed. AIM: To develop and validate clinical and 2D radiomics models for predicting muscle invasion in patients with bladder cancer. METHODS: A retrospective, cross-sectional, multicenter study was conducted. We randomly assigned 80% of the sample to the training set and 20% to the test set. The results of magnetic resonance imaging of the pelvic organs obtained with intravenous contrast according to a standard protocol on tomographs with a magnetic field induction of 1.5 or 3 T were analyzed. All images were processed using a fixed voxel size of 1 × 1 × 1 mm. Clinical and imaging data were analyzed, and texture-based (radiomics) analysis was performed. RESULTS: This study included 84 patients. The median age of the patients was 68.5 years [60.75; 75.0]. The clinical imaging model or predicting muscle invasion was based on 6 parameters: Vesical Imaging-Reporting and Data System (VI-RADS) score of 4 or 5, tumor grade, maximum tumor size, age, VI-RADS score of 1 or 2, and number of tumors. Despite satisfactory specificity (78.6%) and accuracy (72.2%) in the test data, the sensitivity of the model was only 50.0%. The radiomics model included 4 features of muscle invasion selected via the LASSO regression method. The radiomics model demonstrated superior performance against the clinical imaging model, achieving higher accuracy (77.8%) and sensitivity (75.0%) for detecting muscle invasion. CONCLUSION: Texture analysis of magnetic resonance imaging images can be used to differentiate between muscle-invasive and non-muscle-invasive forms of bladder cancer. This study demonstrated the significant scientific and clinical potential of texture analysis for diagnosing muscle invasion in bladder cancer. Our results underscore the potential for further research and substantiate the need for clinical validation.
BACKGROUND: Cognitive load monitoring plays a key role in personalized neurorehabilitation and evaluation of educational program effectiveness. However, existing electroencephalographic systems with a large number of electrodes are technically complex and unsuitable for everyday use. Insufficient studies have explored the minimum electrode configurations necessary for a reliable assessment of the tasks' complexity under cognitive load. AIM: To determine the minimum electroencephalographic electrode configurations that ensure reliable identification of cognitive load levels during multitasking (MATB-II) and working memory (N-Back) tasks, using feature importance and attribution methods, as well as to evaluate the applicability of electrode configurations proposed in previous studies. METHODS: A cross-sectional, single-center study using retrospective data was conducted. Cognitive load was assessed using MATB-II and N-Back tasks from the open COGnitive Brain-Computer Interface dataset. Raw electroencephalographic signals were filtered and subjected to spectral analysis with allocation of power in the θ, α, β and γ bands. The Kolmogorov–Arnold Networks model was used for training and selecting informative features to enable feature importance attribution and determine the minimum electrode configurations. RESULTS: The sns_att4 configuration with four channels and a value of F1 = 0.738 turned out to be the best for the MATB-II task. The sns_gp8 configuration with eight channels and F1 = 0.695 is the best for the N-Back task. Using the Kolmogorov–Arnold Networks model combined with the attribution method enabled the identification of compact electrode configurations (3–10 channels) that maintain high classification accuracy. The optimal 9-electrode configuration (ours_9) for MATB-II provided the mean F1 of 0.714. The three-electrode configuration (ours_3) for N-Back reached F1 = 0.617. Statistical analysis has confirmed that the proposed configurations are competitive with previously published schemes and, in some cases (ours_3 for N-Back), demonstrate an advantage with a minimum number of channels. CONCLUSION: The identified configurations with a limited number of electrodes ensure reliability in determining a cognitive load. They can be used at home setting at the third stage of rehabilitation and integrated into portable neural devices.
BACKGROUND: In the era of low-dose computed tomography replacing chest radiography, attention is increasingly directed not only to lung pathology but also to other organ systems. Early detection of cardiovascular changes and prompt treatment response minimize damage to patient health, potentially reducing mortality and improving both survival and quality of life. The introduction of artificial intelligence into the radiology workflow is considered a significant supportive tool to reduce the omission of clinically important findings and improve the quality of radiology reports. AIM: To compare the diagnostic accuracy of the xAID Chest CT Module artificial intelligence model and radiologists in detecting peripheral solid pulmonary nodules and clinically significant findings, including coronary artery calcification, and pulmonary artery and ascending aortic dilation on routine chest computed tomography. METHODS: This is a retrospective, cross-sectional, single-center study involving 300 chest computed tomography studies were processed by the artificial intelligence model. The reference standard was established by an expert radiologist. Additionally, the respective radiology reports were reviewed to compare the diagnostic accuracy of the artificial intelligence model and the radiologists in detecting peripheral pulmonary nodules, coronary calcification, and pathological dilation of the ascending aorta and pulmonary trunk. RESULTS: For peripheral pulmonary nodules ≥ 6 mm, the xAID Chest CT Module demonstrated a sensitivity of 81.6% and a specificity of 91.0%, whereas routine radiology reports demonstrated 100% diagnostic accuracy in detecting pulmonary nodules. For coronary artery calcification, the artificial intelligence demonstrated a sensitivity of 97.0% and a specificity of 50.0%, compared with a radiologist sensitivity of 79.8% and a specificity of 100%. For main pulmonary artery dilation ≥ 29 mm, artificial intelligence sensitivity and specificity were 95.8% and 78.1%, respectively, whereas the radiologist achieved a sensitivity of 56.9% and a specificity of 100%. For ascending aortic dilation ≥ 40 mm, artificial intelligence reached sensitivity of 100% and specificity of 98.9%, compared with a radiologist sensitivity of 66.7% and a specificity of 100%. CONCLUSION: The xAID Chest CT Module cannot be considered a standalone tool for diagnosing peripheral pulmonary nodules and coronary calcification; rather, it should be regarded as a support tool in the radiologist workflow to draw attention to small nodules and manifestations of coronary calcification. However, the model's high diagnostic accuracy in assessing the diameters of the ascending aorta and pulmonary trunk offers the potential of a standalone tool for opportunistic screening of vascular pathology using chest computed tomography data.
BACKGROUND: Developing highly sensitive diagnostic methods for common extrapyramidal disorders such as Parkinson's disease and essential tremor is a pressing research challenge. Despite the rapid development of various diagnostic methods, including functional radioisotope neuroimaging, differential diagnosis of Parkinson's disease and essential tremor is often challenging. Electrooculographic signals provide additional information that can be leveraged to develop new approaches to the differential diagnosis of Parkinson's disease and essential tremor. AIM: To identify specific quantitative characteristics of saccadic eye movements in Parkinson's disease and essential tremor patients using a new statistical analysis method for biomedical signals. METHODS: An observational, cross-sectional, single-center study with retrospective data analysis was conducted. Outpatients were divided into two groups: Group 1, patients with Parkinson's disease; and Group 2, patients with essential tremor. During electrooculogram, patients were presented with a visual stimulus on a monitor: a white square on a black background moving from bottom to top within a 15°visual field. The stimulus was presented eight times, each presentation lasting three seconds. Electrooculographic signals were analyzed in their raw form, without prior signal segmentation. Statistical analysis was performed using macrosaccade analysis software and burst electrical activity analysis software with ROC-AUC diagrams. RESULTS: Group 1 included 17 patients in stage 1–3 Parkinson's disease, and Group 2 included 10 patients with essential tremor. A decrease in electrooculographic macrosaccade latency was observed in both Group 1 and Group 2 in left and right eyes during the presentation of a series of the visual stimuli (Kendall's τb, p ≤ 0.031). The analysis of electrooculographic micromovements revealed wave trains that differentiated Parkinson's disease and essential tremor patient groups (two-tailed Brunel–Munzel test; p ≤ 0.03 for both eyes). A correlation was found between the number of spikes in electrooculographic of the left and right eye in Parkinson's disease patients (Kendall's τb is 0.515; p = 0.011). CONCLUSION: The new method of statistical signal analysis enabled the identification of new quantitative features in electrooculographic in Parkinson's disease and essential tremor patients. These new features were found in both macrosaccade and eye micromovement parameters. The identified features are offer potential for new methods for differential diagnosis of Parkinson's disease and essential tremor.
Over the past decades, detection rate of renal tumors has increased. However, some of them, especially at the T1a stage, might be benign. Radiology methods are often unable to reliably determine a tumor's nature, so histological verification remains the gold standard. The use of noninvasive methods for diagnosing tumors is particularly relevant in modern urology. Radiomics is a promising direction, allowing for the prediction of histological verification results by quantitative assessment of intratumoral heterogeneity. However, a lack of standardization and external validation of modern radiomics methods hinders their widespread implementation. Over the past few years, radiomics analysis has been increasingly integrating machine learning methods, which reduce the risk of bias. This article presents a review of publications devoted to radiomics analysis of magnetic resonance and computed tomography images in the differential diagnosis of angiomyolipomas without macroscopically detectable fat and other types of minimal-fat renal tumors. We searched for Russian publications at the eLibrary system; however, the given keywords did not yield relevant results. Therefore, we searched non-Russian literature using PubMed, MEDLINE and Embase databases, as well as ClinicalTrials.gov registry published between January 2021 and 2025 that addressed radiomics analysis of magnetic resonance and computed tomography images in the differential diagnosis of angiomyolipomas without macroscopically detectable fat and other renal tumors. We initially selected 792 articles. After applying the eligibility criteria and excluding duplicates, we included nine articles in the review. Each study included in the review demonstrated a high Area Under the Curve for the predictive model. Second-order features were identified as statistically significant in 67% of cases. Machine learning methods were used in each study. The quality of the studies, assessed based on their methodologies, was considered satisfactory. Its decrease was primarily due to the lack of external validation of the results (89% of the analyzed articles), as well as the data being unavailable or lacking transparency (82%). The reproducibility of the radiomic parameters can be considered low. Consequently, strict standardization and the availability of open radiomics datasets are essential for the clinical adoption. The integration of machine learning algorithms in radiomics can improve sensitivity, specificity, and accuracy in the differential diagnosis of renal tumors.
BACKGROUND: Suboptimal quality of chest radiographs is a common problem in the diagnostic imaging of thoracic diseases. Quality assessment of routine chest radiographs is performed manually, which is complicated by remote reporting and the high workload of radiologists. We previously developed an artificial intelligence-powered tool for quality control of these images. However, its performance in real-world clinical practice has not been investigated. AIM: To evaluate the effectiveness of the automated artificial intelligence tool for detecting defects in chest radiographs in routine clinical practice. METHODS: The tool was validated using data from three Moscow outpatient hospitals over a two-month period. Chest radiographs were assessed using an automated quality assessment tool based on ensemble learning that sequentially processes chest radiographs in DICOM (Digital Imaging and Communications in Medicine) format. The tool is integrated into the routine chest radiography workflow at the Moscow Reference Center for Radiology. Following automated assessment, image quality was analyzed by radiologists (n = 10). Based on the automated assessment, a report was generated that included information on incomplete visualization of the lungs and costodiaphragmatic recess in frontal and lateral view, chest rotation, incorrect image orientation, and incorrect information (metadata) about the anatomical region, view, and photometric interpretation of the study. Radiologists assessed radiographic image defects using similar criteria. Each image was evaluated by a single radiologist. The radiologists were blinded to the automated image evaluation results. RESULTS: A total of 9642 radiographs were processed. The mean processing time per image was 14 seconds. In total, 3386 distinct patient positioning errors were identified, with each error detected in 14%–45% of images. The accuracy of metadata completion for radiographs ranged from 14% to 100%. The accuracy, expressed as the ROC AUC, for detecting patient positioning errors using the quality control tool ranged from 0.782 to 0.947, while the accuracy for detecting metadata completion errors ranged from 0.985 to 1.0. CONCLUSION: Automated quality control of radiographs is an accurate and rapid method for detecting patient positioning errors and errors in metadata completion.
BACKGROUND: In patients with stage 5 chronic kidney disease receiving maintenance hemodialysis, the assessment of bone mineral density is of special clinical importance due to the high prevalence of mineral and bone disorders and low-energy fractures. However, interpreting dual-energy X-ray absorptiometry findings in this population is challenging because measurements obtained from different anatomical sites vary in clinical utility. Additional limitations exist when utilizing the FRAX algorithm. AIM: To determine the most diagnostically informative anatomical sites for bone mineral density measurement using dual-energy X-ray absorptiometry in patients with stage 5 chronic kidney disease receiving maintenance hemodialysis, and to evaluate the limitations of the FRAX tool in this population. METHODS: A prospective, single-center, cross-sectional analytical study was conducted. The primary cohort included 32 patients with stage 5 chronic kidney disease receiving maintenance hemodialysis, while the control cohort consisted of 20 individuals without chronic kidney disease. Statistical analysis included a comparative assessment of densitometric parameters across distinct anatomical sites, distribution of patients according to World Health Organization diagnostic categories, Z-scores, FRAX values, and the correlation between densitometric parameters and serum parathyroid hormone levels. RESULTS: In the between-group comparison, femoral neck T-scores were significantly lower in the primary cohort than in the control group 1.63 ± 0.41 vs. 0.07 ± 0.70 (p 0.001). After adjusting for age and sex, assignment to the primary cohort remained strongly associated with lower femoral neck T-scores [β = 1.60 (95% CI − 2.06… − 1.14), p 0.001]. T-scores at the Total Hip site were also significantly lower in the primary cohort (−1.28 ± 0.72 vs. 0.08 ± 0.86 (p = 0.001)). Conversely, for the lumbar spine, between-group differences did not reach statistical significance (p = 0.111). Osteopenia at the femoral neck was identified in 15 of 17 patients in the primary cohort (88.2%) but was not detected in the control group. The femoral neck Z-scores were significantly lower in the primary cohort than in the controls [−1.59 ± 1.02 vs. 0.41 ± 0.83 (p 0.001)], demonstrating a more pronounced reduction in bone density relative to age- and sex-matched norms. Correlation analysis revealed a strong trend toward an inverse association between femoral neck T-scores and serum parathyroid hormone levels (ρ = −0.56; p = 0.058; n = 12). While the primary cohort demonstrated higher FRAX values compared to controls, a wide overlapping range persisted between the two groups. CONCLUSION: In patients with stage 5 chronic kidney disease receiving maintenance hemodialysis, proximal femur densitometric parameters — specifically at the femoral neck — demonstrated superior diagnostic utility compared to lumbar spine measurements. In this cohort, FRAX values should be viewed as a complementary tool.
The prevalence of thyroid nodules is relatively high, affecting up to 68% of the adult population. Although most nodules are benign and do not require treatment, therapy may be needed in some cases when they cause symptoms such as compression, cosmetic concerns, or hyperthyroidism. Minimally invasive techniques, particularly ultrasound-guided radiofrequency ablation (RFA), are gradually assuming an important role in the treatment of carefully selected patients, demonstrating a favorable efficacy and safety profile. Given the high operator dependence of ultrasound (US) and the technical complexity of RFA, growing interest has emerged in the implementation of objective artificial intelligence (AI)-based tools designed to support clinical decision-making at all stages of the procedure. Objective: to assess the potential applications of artificial intelligence in radiofrequency ablation of thyroid nodules. A systematized literature search was conducted in PubMed/MEDLINE, Embase, Web of Science, Scopus, IEEE Xplore, CNKI, Wanfang, CiNii, KoreaMed, KISS, and DBpia. The review included studies, clinical reports, and technical developments addressing the use of machine learning and computer vision methods in radiofrequency ablation for thyroid nodules. In preoperative diagnostics, artificial intelligence systems, according to meta-analyses, demonstrate high diagnostic performance comparable to that of expert physicians and contribute to reducing the number of unnecessary invasive procedures. Predictive models of radiofrequency ablation efficacy integrating clinical and ultrasound predictors have also been developed. For intraoperative support, novel ultrasound navigation technologies have been proposed. In the post-ablation period, the main challenge remains the accurate assessment of ablation completeness. Algorithmic approaches and automated methods for measuring the ablation zone have been developed; however, clinically validated models for segmentation of post-ablation changes in thyroid nodules remain limited. Conclusion: artificial intelligence technologies demonstrate the highest degree of clinical applicability at the stage of preoperative diagnostics. At the same time, solutions for the intraoperative and postoperative stages require further research, confirmation of their impact on long-term clinical outcomes, and integration into routine clinical practice.
This review is devoted to the application of neural network models for the analysis of histological images in non-neoplastic liver diseases, with a focus on the technical aspects of model development and training. The relevance of this work is due to the growing interest in applying neural networks and computer vision for research and diagnostic tasks related to microscopic morphology. Although this disease group does not lose its clinical significance, and the importance of microscopic verification of morphological changes, the application of artificial intelligence in this field remains limited and fragmented. The review systematizes available annotated datasets of liver histological images, applied neural network architectures, image preprocessing approaches, and training strategies. It also examines the loss functions used and other key technical aspects of neural network development. It is shown that, although the contribution of such models to the automation and standardization of the morphological assessment appears promising, their practical implementation is constrained by the limited publicly available annotated data, the high labor intensity of annotation, and insufficient standardization of methodological approaches. It is noted that multimodal data, integrating histological images with clinical, biochemical, or radiological parameters, is of great clinical interest; however, it is currently rare. Most studies utilize universal neural network architectures, while models fine-tuned to work with microscopic features and liver morphology are applied much less commonly. The analysis of published studies has shown that weakly supervised learning can be sufficient for model development and, in combination with significantly lower annotation costs, offers substantial potential for further development. At the same time, a considerable proportion of studies describe methodologies with limited reproducibility.
Over the past decades, the detection rate of kidney tumors has increased. However, some of them may be benign. Radiographic diagnostic methods are often unable to reliably determine the tumor's nature, so histological verification remains the gold standard. The use of noninvasive methods for diagnosing tumors is particularly relevant in modern urology. One such promising method is radiomics, which is based on mathematical methods for assessing tumor structure heterogeneity and allows for the prediction of histological verification results. Nevertheless, the lack of standardization and external validation of modern radiomics methods hinders their widespread implementation. Over the past few years, radiomics analysis has increasingly been used in conjunction with machine learning (ML) methods, which reduce the risk of systematic errors. This paper presents a review of publications devoted to MR and CT radiomics in differentiating fat-poor AML from other types of kidney tumors with low lipid content. It provides information on the role of radiomics in determining the nature of kidney tumors, details the methodology for performing radiomics analysis, and demonstrates the results of radiomics analysis for the differential diagnosis of fat-poor AML from other renal tumors based on CT and MR images, identifying the most reliable textural features. A search of publications using keywords in the Russian scientific database eLIBRARY.RUyielded no results. Therefore, a search of English-language literature was conducted using PubMed/MEDLINE, Embase, and the ClinicalTrials.gov registry, published between January 2021 and 2025, which used MR and CT radiomics to differentiate BG-AML from other renal tumors. A total of 791 articles were initially identified. After applying eligibility criteria and excluding duplicate publications, nine articles were included in the review. Each study included in the review demonstrated high AUC values for the predictive model. The combination of radiomics with clinical and/or conventional imaging features outperformed both clinical and radiomics only models. Second-order features (GLCM, GLRLM, GLSZM, GLDM, NGTDM) were identified as statistically significant in 67% of cases. Machine learning methods were used in every study. Logistic regression was the most common machine learning algorithm (7/9 studies). Radiomics features were extracted from CT images in 6/9 studies, while MRI-based radiomics was used in three studies. Most studies (6/9) focused on differentiating fp-AML from clear cell renal cell carcinoma. The quality of the studies, assessed based on their methodology, was considered satisfactory. This was primarily due to the lack of external validation of results (89% of the included articles), as well as the unavailability or opacity of data (82%). It is essential to highlight that there is no strict standardization in radiomics analysis. Some authors used 2D segmentation (in 5/9 studies) and manual region of interest (ROI) delineation in most studies.Thus, the reproducibility of radiomics parameters can be considered low. All included studies were retrospective, corresponding to the current state of radiomics research. Differentiating fp-AML remains challenging due to its low prevalence, which leads to small sample sizes and class imbalance. Therefore, strict standardization and the creation of an open radiomics database are necessary for the implementation of radiomics in clinical practice. The use of machine translation algorithms in radiomics can lead to increased sensitivity, specificity, and accuracy in distinguishing renal masses. Further prospective multicenter studies with external validation are needed to introduce radiomics into clinical practice.
Air embolism is considered a rare but potentially fatal complication of invasive thoracic surgery. Computed tomography-guided thoracic biopsy is widely employed in clinical practice as a highly effective and relatively safe method for the morphological profiling of lung tumors. However, some publications indicate the possibility of systemic air embolism leading to subsequent severe cardiovascular and neurological complications. At present, predisposing factors, the pathophysiological mechanisms underlying the systemic embolism and the optimal acute-phase management remain poorly understood, making clinical observations of scientific and practical value. In this case report, a 75-year-old patient developed systemic air embolism after computed tomography-guided transthoracic needle biopsy of a right upper-lobe tumor, which subsequently resulted in acute myocardial infarction and acute ischemic cerebrovascular accident. The patient received conservative therapy in the intensive care unit and was discharged with a neurological deficit for rehabilitation. Histological and immunohistochemical tests confirmed lung adenocarcinoma. Six months thereafter, radical surgical intervention — a right upper lobectomy — was performed. This case underscores the need for careful patient selection for invasive diagnostics, mandatory early post-procedure monitoring, and readiness of medical staff to promptly recognize and manage rare complications. It broadens the clinical spectrum of systemic air embolism and emphasizes the importance of further investigation in modern thoracic oncology.
BACKGROUND: Concomitant thyroid disorders can decrease the diagnostic accuracy of topical imaging modalities used to localize primary hyperparathyroidism. Therefore, in such cases, the effectiveness of radionuclide imaging for parathyroid lesions should be confirmed. AIM: This study aimed to evaluate the effects of nodular and autoimmune thyroid disorders on the diagnostic accuracy of radionuclide imaging for parathyroid lesions in patients with primary hyperparathyroidism and compare the accuracy of radionuclide imaging with that of other imaging modalities. METHODS: The study included three patient groups: patients with primary hyperparathyroidism without a concomitant thyroid disease (group 1; n = 50), patients with autoimmune thyroid disorder (group 2; n = 50), and patients with thyroid nodule/multinodular goiter (group 3; n = 50). All the patients underwent ultrasound, planar scintigraphy, and single-photon emission computed tomography with X-ray computed tomography prior to parathyroidectomy. If negative or equivocal results were obtained (e.g., suspected parathyroid lesion), a contrast-enhanced computed tomography scan was performed. The intervals between scans and between the first scan and parathyroidectomy did not exceed 6 months and 12 months, respectively. Sensitivity and the positive predictive value were estimated. RESULTS: In group 2, the diagnostic accuracy of radionuclide imaging for parathyroid lesions was lower than that of other imaging modalities. Moreover, in group 2, contrast-enhanced computed tomography demonstrated a diagnostic sensitivity of 79%, and its combination with ultrasound achieved a higher sensitivity of 85%. In group 3, the highest diagnostic value was obtained for single-photon emission computed tomography combined with X-ray computed tomography (85%) or ultrasound (88%). The results of radionuclide imaging of parathyroid lesions in patients with primary hyperparathyroidism and autoimmune thyroid disease depended on the volume, density, and vascularization of the thyroid gland. CONCLUSION: Single-photon emission computed tomography combined with X-ray computed tomography is an accurate diagnostic modality for identifying parathyroid lesions in patients with primary hyperparathyroidism and concomitant nodular thyroid disorder. Contrast-enhanced computed tomography showed the highest sensitivity in patients with primary hyperparathyroidism and autoimmune thyroid disease.
BACKGROUND: Proper fecal tagging allows for high-quality computed tomography colonography. However, there is no single tagging scheme. Therefore, the effects of a contrast enhancement regimen on fecal tagging should be evaluated. AIM: This study aimed to compare the quality of single-dose fecal tagging with that of split-dose fecal tagging with iohexol during computed tomographic colonography and to assess the impact of these regimens on procedure tolerability. METHODS: In this retrospective, selective, single-center study, the patients were divided into two groups based on whether they received single-dose (group 1) or split-dose (group 2) fecal tagging. Both groups received 50 mL of the iodine-containing contrast agent iohexol, with iodine concentration of 350 mg/mL. The residual liquid density was assessed using three parameters: maximum, minimum, and mean values. Additionally, the residual fluid homogeneity was assessed by calculating the mean standard deviation within the region of interest. Tolerability of preparation for colonography was assessed using a 10-point visual analog scale. RESULTS: The final sample included 338 patients: 116 in group 1 and 222 in group 2. The mean, minimum, and maximum density values in group 2 were significantly higher than those in group 1: 943 [722; 1245], 753 [525; 1082], and 1079 HU [801; 1456] versus 681 [420; 907], 570 [374; 820], and 825 HU [496; 1154], respectively (p 0.001). The residual fluid homogeneity was significantly higher in group 2 than in group 1: 59 [46; 78] versus 67 HU [54; 81] (р = 0.012). Group 2 showed a significantly lower subjective difficulty of preparation than did group 1: 4 [2; 6] and 5 [4; 7], respectively (p = 0.004). CONCLUSION: A single dose of 50 mL of iohexol (iodine concentration: 350 mg/mL) provides higher-quality fecal tagging than a split-dose provides because of higher residual fluid density with maintained homogeneity. Moreover, single-dose tagging was found to be more tolerable.
This article explored the role of pareidolia in radiography and its potential in improving diagnosis and medical personnel training. Pareidolia is the phenomenon of perceiving familiar patterns in random objects, such as faces on the moon’s surface and animal figures in clouds. In radiography, pareidolia can manifest as recognizable patterns in medical images. This enables radiographers to identify abnormalities and improve their diagnostic skills. This work aimed to evaluate pareidolia caused by the interpretation of X-ray images and determine its potential applications. From June to December 2023, a competition was held to create a dataset of pareidolic illusions. Thirty-one individuals participated, including medical imaging specialists who had access to radiographic images. Images from nine additional participants were obtained outside the competition. Overall, 71 images were received. Participants uploaded images using a form on Yandex Forms. Data quality was ensured by clearly defined inclusion and exclusion criteria. Data analysis revealed that people most frequently perceive human faces, animal snouts, and the heart symbol. These findings indicate the possibility of further research. This article discusses the potential applications of pareidolia in developing neural networks for automated medical image analysis and in educational activities that stimulate creative thinking and association. Moreover, the article emphasizes the importance of ongoing research in this area to develop effective diagnostic tools and educational programs by expanding the evidence base.
BACKGROUND: Standard magnetic resonance imaging sequences only provide qualitative image assessment, which is rather subjective. However, some quantitative techniques can interpret findings more objectively and expand diagnostic capabilities. Previously, they were mainly used for brain and joint scans; however, current technology allows using them for evaluating peripheral nerve function. AIM: This study aimed to evaluate quantitative parameters of magnetic resonance imaging of the brachial plexus elements in healthy adults, depending on the side and level of spinal nerves and demographic and anthropometric characteristics. METHODS: Ten healthy volunteers were included. Their main demographic and anthropometric characteristics were recorded before they underwent magnetic resonance imaging. A 3T magnetic resonance imaging scanner was used. In addition to standard sequences, the scan protocol included regimens for obtaining T2 relaxation times and magnetization transfer ratios from nerve elements of the brachial plexus. Data were post-processed using the MATLAB software package. Then, regions of interest were manually assigned to in the maps, and numerical values were obtained. Furthermore, thickness of the nerve elements was measured. Data were statistically processed using the SPSS software. RESULTS: In each participant, the numerical values of the quantitative magnetic resonance imaging parameters (measured T2 relaxation time, proton density, magnetization transfer ratio, and thickness) in the anterior rami of the spinal nerves that form the brachial plexus were obtained. The thickness gradient of the normal anterior rami was revealed, with the highest value occurring at the level of the anterior rami of cervical spinal nerve C7. Significant positive correlations between T2 relaxation time and age were determined by analysis of the associations between quantitative magnetic resonance imaging parameters and demographic and anthropometric characteristics. In addition, negative correlations were found between height and measured T2 relaxation time and proton density. CONCLUSION: The study results indicate that future research on T2 relaxation parameters should consider age and height in both healthy volunteers and patients with a brachial plexus condition. Additionally, when measuring the thickness of the anterior rami of the brachial plexus using standard sequences, the size and thickness gradient of the nerve elements should be considered.
BACKGROUND: Photoplethysmography, a method used to measure blood volume changes per pulse, is widely applied in healthcare. In Persian medicine, pulsology is considered one of the most important methods for clinical diagnosis. However, recently the theory of fuzzy sets has provided a valuable foundation for developing knowledge-based systems in medical research. AIM: To estimate and predict the systolic area of photoplethysmography signals using Persian medicine pulsology, by leveraging the potential of fuzzy systems. METHODS: To design the fuzzy controller, a Persian medicine specialist simultaneously recorded data on PM pulse characteristics including pulse frequency and pulse strength—along with photoplethysmography signals, from 55 healthy volunteers. Initially, rules were generated based on the input and output variables. After evaluating these rules using the collected data, 35 were retained and presented in a two-input–one-output lookup table. RESULTS: The fuzzy system was then constructed using MATLAB. It included 35 rules, triangular and trapezoidal membership functions, a singleton fuzzifier, a product inference engine, and a center-average defuzzifier. This system, which used pulse frequency and pulse strength as inputs and provided the systolic area as output, demonstrated acceptable performance within the defined input range. CONCLUSIONS: The proposed fuzzy controller system reasonably predicted the systolic area of photoplethysmography signals using Persian medicine pulse parameters. The results revealed that increasing pulse frequency decreased the systolic area, while increasing pulse strength increased it, in alignment with previous results. Therefore, this system may boost the clinical skills of Persian medicine students and practitioners. It also holds promise for application in disease diagnosis and prediction and for facilitating integration between Persian medicine and mainstream medicine.
BACKGROUND: Dental cone-beam computed tomography offers several advantages, including superior image quality, an acceptable size, and a lower radiation dose compared to conventional CT scanning. Moreover, cone-beam computed tomography is more suitable for dentists to acquire and analyze images, and it provides greater comfort for patients due to technological advancements. Cone-beam computed tomography generates three-dimensional images of the head and neck and is utilized across various dental fields, including dental surgery, endodontics, trauma, implant dentistry, head and neck lesions and diseases, and orthodontics. AIM: To evaluate seven tissue doses using three scan protocols on a cone-beam computed tomography scanner KaVo OP 3D Pro, and to investigate the effect of resolution options on the effective dose. METHODS: Three protocols were employed in this study. Three voxel size settings were assessed: 420 μm, 380 μm, and 320 μm. The field of view and tube voltage were kept constant at 13 cm × 15 cm and 90 kV, respectively. Other were, scan time (8–27 seconds) and dose range (50–350 μSv). The absorbed and effective doses were calculated for each cone-beam computed tomography scan protocol. RESULTS: In the throat, the highest dose was absorbed by tissue-2 (7.719 mGy). In the teeth, the highest dose was absorbed by tissue-3 (16.326 mGy). In the cheek, the maximum dose was absorbed by tissue-3 (25.053 mGy). In the eyes, the highest dose was absorbed by tissue-3 (12.962 mGy). In the forehead, the highest dose was absorbed by tissue3 (8.465 mGy). In the mid-skull, the highest dose was absorbed by tissue-3 (20.904 mGy). In the occipital region, the highest dose was absorbed by tissue-2 (7.8 mGy). Regarding effective doses, protocol-3 generally resulted in higher values, except in the throat and occipital regions, where tissue-2 absorbed more. CONCLUSION: This study demonstrates that changes in cone-beam computed tomography exposure parameters influence the effective dose. Adjusting resolution settings results in variations in effective doses, highlighting the significance of selecting appropriate exposure factors, such as voxel size or resolution options. Dentists must carefully consider imaging parameters, as these decisions have a direct impact on patient exposure.
BACKGROUND: Texture analysis improves the diagnostic accuracy of magnetic resonance imaging and differential diagnosis of prostate lesions, which are primarily segmented through manual labeling, resulting in significant inter-expert variability of masks. A consensus-based technique can help reduce inconsistencies in prostate lesion segmentation. However, global scientific studies have not described any standardized, consensus-based labeling protocols. AIM: This study aimed to develop a consensus algorithm for manual labeling of prostate lesions by several independent experts and evaluate inter-expert consistency in lesion segmentation. METHODS: This retrospective study included 60 biparametric magnetic resonance imaging scans of the prostate gland performed according to PI-RADS 2.1 technical specification. The scans showed PI-RADS 3, 4, and 5 lesions. Two independent radiologists manually segmented the prostate lesions using 3D Slicer. Then, the resulting masks were compared using the Dice–Sørensen coefficient (DSC). For lesions with DSC ≥ 0.75, the final mask was based on the overlap between the two original masks. Conversely, for lesions with DSC 0.75, the final mask was determined using the proposed consensus algorithm. RESULTS: The proposed consensus algorithm significantly increased the DSC values, from 0.61 [0.48; 0.73] for primary labeling to 0.74 [0.62; 0.79] for labeling using the proposed algorithm (p = 0.01). CONCLUSION: The proposed consensus-based algorithm for labeling prostate lesions using magnetic resonance imaging data is crucial in addressing inadequate approaches to objective segmentation in research and clinical settings.
BACKGROUND: Patient examinations generate large amounts of data on various diseases, which are virtually impossible to process manually. Currently, automated radiology workstations incorporate clinical decision–support systems that facilitate image analysis. At present, two options are used: built-in vendor-dependent and vendor-independent software solutions. Both have broad functionality, but they also have their advantages and drawbacks. Therefore, the choice of the best clinical decision–support system depends on the healthcare organization. AIM: This study aimed to compare two approaches for selecting radiodiagnostic solutions for automated radiology workstations in Moscow. METHODS: The study conducted a two-stage, cross-sectional survey between September 2023 and March 2024. Data on vendor-independent software were obtained from participants of the Experiment on the use of innovative computer vision technologies for analysis of medical images in the Moscow healthcare system. Conversely, data on vendor-dependent solutions was collected from the manufacturers’ websites. Furthermore, a survey of 40 radiologists was performed to evaluate the relevance of software functions. RESULTS: At the time of the survey, the vendor-independent software demonstrated slightly more functions than the vendor-dependent solutions. The greatest differences were observed in the functionality of computed tomography and magnetic resonance imaging modalities, whereas the functionality for X-ray imaging and mammography was nearly identical. The survey of radiologists showed that 6 of 17 functions were unique to built-in vendor-dependent software. However, 40% of the radiologists actually needed these functions, whereas the shared functions were relevant for 50% of the respondents. CONCLUSION: Vendor-independent and built-in vendor-dependent software solutions share only half of their functions. Thus, in choosing between these two options, healthcare facilities should consider their specialization and their radiologists’ requests. Moreover, approximately two-thirds of the built-in vendor-dependent software functions used by radiologists in Moscow can be implemented using vendor-independent solutions. Therefore, the choice of software should be based on the facility’s technical capacity and economic feasibility.
