Optimal time-frequency localized wavelet filters for identification of Alzheimer's disease from EEG signals

Alzheimer's disease (AD) is a chronic disability that occurs due to the loss of neurons. The traditional methods to detect AD involve questionnaires and expensive neuro-imaging tests, which are time-consuming, subjective, and inconvenient to the target population. To overcome these limitations, Electroencephalogram (EEG) based methods have been developed to classify AD patients from normal controlled (NC) and mild cognitive impairment (MCI) subjects. Most of the EEG-based methods involved entropy-based feature extraction and discrete wavelet transform. However, the existing AD classification methods failed to provide promising classification accuracy. Here, we proposed a wavelet-machine learning (ML) framework to detect AD using a newly designed biorthogonal filter bank by optimization of frequency and time localization of triplet halfband filter banks (OTFL-THFB). The OTFL-THFB decomposes EEG signals into various EEG sub- bands. Hjorth Parameters (HP) and Higuchi's Fractal Dimension (HFD) have been investigated to extract features from each EEG subband. Subsequently, ML models are trained and tested using different features such as OTFL-THFB with HFD, OTFL-THFB with HP, and OTFL-THFB with HFD and HP used for detecting AD with 10-fold cross-validation. This method was applied to two publicly available datasets. Our model achieved an accuracy of 98.91 % for AD versus NC and 98.65 % for AD versus MCI versus NC using the least square support vector machine. Results indicate that this framework surpassed existing state-of-the-art techniques for classifying AD from NC.

Biomechanical properties of porcine cornea; planar biaxial tests versus uniaxial tensile tests

The cornea is a transparent tissue whose mechanical properties are important for its optical and physiological functions. The mechanical properties of cornea depend on the composition and microstructure of its extracellular matrix, which is composed of collagen fibrils with preferential orientations. The present research was done in order to characterize corneal mechanical response using the biaxial mechanical testing method and to compare biaxial measurements with those found from uniaxial tensile tests. For this purpose, thirty square-shaped specimens excised from the center of porcine cornea were mounted into an ElectroForce TestBench device such that their superior/inferior (SI) and nasal/temporal (NT) meridians were aligned with motor axes. Furthermore, ten corneal strips dissected from the NT direction (n = 5) and SI direction (n = 5) were mounted into an RSA-G2 Solid Analyzer testing machine. The biaxial experiments were performed at stretch ratios of 1:1, 1:0.5, 0.5:1, 1:0.01, and 0.01:1 and displacement rates of 2 mm/min (n = 20) and 10 mm/min (n = 10). The uniaxial experiments were done using the displacement rate of 2 mm/min. The planar square-shaped samples tested under equibiaxial loading showed similar mechanical response in NT and SI directions. Furthermore, uniaxial experiments revealed no significant difference in tensile response of corneal strips excised from NT and SI directions. However, equibiaxial testing tensile stresses were significantly larger than those found from uniaxial tensile measurements. The mechanical behavior of cornea in biaxial tests was dependent on the applied stretch ratio. The differences and similarities between uniaxial and biaxial experimental measurements were discussed and it was concluded that the planar biaxial testing method characterized the mechanical response of cornea by mimicking its in vivo loading state more closely than uniaxial experiments.

Evaluation of flavor perception of strong-aroma Baijiu based on electroencephalography (EEG) and surface electromyography (EMG) techniques

This study investigates the flavor perception of strong-aroma Baijiu through physiological electrical signals, focusing on electroencephalography (EEG) and electromyography (EMG) during olfactory and gustatory evaluations. It examines how sensory qualities, especially mellowness, influence brain and muscle responses. Results showed significant differences in EEG δ and β wavebands, mainly in the frontal and temporal lobes, reflecting varying brain activities across Baijiu types. Mellower Baijiu triggered fewer activations in the digastric muscle, indicating a smoother swallowing experience. Baijiu with higher sensory scores reduced corrugator muscle activity and increased zygomatic major muscle activation, indicating a more pleasant taste. The findings highlight that Baijiu flavor perception is shaped by memory, emotion, and sensory inputs, demonstrating its complexity. This study offers valuable insights into flavor perception's multimodal nature and suggests novel ways to refine Baijiu evaluation using physiological signals to understand taste quality.

FLASH Radiotherapy: From In Vivo Data to Clinical Translation

Delivery of radiotherapy (RT) at ultra-high dose rates or FLASH radiotherapy (FLASH-RT) is an emerging treatment option for patients with cancer that could increase survival outcomes and quality of life. In vivo data across a multitude of normal tissues and associated tumors have been published demonstrating the FLASH effect while bringing attention to the need for additional research. Combination of FLASH-RT with other treatment options including spatially fractionated RT, immunotherapy, and usage in the setting of reirradiation could also provide additional benefit. Phase I clinical trials have shown promising results, yet research is warranted before routine clinical use of FLASH-RT.

Evaluation of 18F-FDG absorbed dose ratios in percent in adult and pediatric reference phantoms using DoseCalcs Monte Carlo platform

This study investigates the field of radiation exposure in nuclear medicine, which might have implications for exposure to ionizing radiation in pediatric cases. To demonstrate the difference in radiosensitivity between younger patients and adults and to highlight the need for individualized radiation protection procedures when investigating medical imaging and therapy, this study examines the absorbed dose ratios in percent (ADR%) for 18F-FDG. This parameter is an important indicator, illustrating the percentage of radiation dose absorbed by specific organs/tissues concerning the emitted radiation from different body regions. The methodology involves calculating ADR% in twelve voxel-based models for adults, children, and newborns, as referenced by International Commission on Radiological Protection (ICRP) Publications 110 and 143. The simulations used the 18F positron spectrum from ICRP Publication 107 and Livermore models. These simulations were performed using the DoseCalcs Monte Carlo platform. We have calculated the S-values and ADR% using the DoseCalcs simulations of the 18F positrons and provided a comprehensive dataset of ADR% results. This dataset evaluates the impact of anatomical variation on absorbed dose in target regions. It consists of 141 target regions and 8 different source regions. Significant differences in radiosensitivity were observed in ADR% values among various source-target combinations for each age and sex group. The self-irradiation ADR% reaches up to 95%, while the cross-irradiation ADR% varies, ranging approximately from 0.1% to 12%, depending on the mass of the target organ, the distance between it and the source organ, and the chemical composition of these organs. Also, the variations observed across different age and sex phantoms highlight the importance of personalized internal dosimetry, especially for pediatric cases with heightened radiosensitivity. Healthcare practitioners can use the dataset of ADR% values as the first stage to illustrate variability and optimize nuclear medicine imaging with 18F-FDG while reducing radiation risks.

Using deep learning for ultrasound images to diagnose chronic lateral ankle instability with high accuracy

The purpose of this study is to calculate diagnostic accuracy of chronic lateral ankle instability (CLAI) from a confusion matrix using deep learning (DL) on ultrasound images of anterior talofibular ligament (ATFL). The study included 30 ankles with no history of ankle sprains (control group), and 30 ankles diagnosed with CLAI (injury group). A total of 2000 images were prepared for each group by capturing ultrasound videos visualizing the fibers of ATFL under the anterior drawer stress. The images of 20 feet in each group were randomly selected and used for training data and the images of remaining 10 feet in each group were used as test data. Transfer learning was performed using 3 pretraining DL models, and the accuracy, precision, recall (sensitivity), specificity, F-measure, and the area under the receiver operating characteristic curve (AUC) were calculated based on the confusion matrix. The important features were visualized using occlusion sensitivity, a method for visualizing areas that are important for model prediction. DL was able to diagnose CLAI using ultrasound imaging with very high accuracy and AUC in three different learning models. In visualization of the region of interest, AI focused on the substance of the ATFL and its attachment on the fibula for the diagnosis of CLAI.

Subtraction-free artifact-aware digital subtraction angiography image generation for head and neck vessels from motion data

Digital subtraction angiography (DSA) is an essential diagnostic tool for analyzing and diagnosing vascular diseases. However, DSA imaging techniques based on subtraction are prone to artifacts due to misalignments between mask and contrast images caused by inevitable patient movements, hindering accurate vessel identification and surgical treatment. While various registration-based algorithms aim to correct these misalignments, they often fall short in efficiency and effectiveness. Recent deep learning (DL)-based studies aim to generate synthetic DSA images directly from contrast images, free of subtraction. However, these methods typically require clean, motion-free training data, which is challenging to acquire in clinical settings. As a result, existing DSA images often contain motion-affected artifacts, complicating the development of models for generating artifact-free images. In this work, we propose an innovative Artifact-aware DSA image generation method (AaDSA) that utilizes solely motion data to produce artifact-free DSA images without subtraction. Our method employs a Gradient Field Transformation (GFT)-based technique to create an artifact mask that identifies artifact regions in DSA images with minimal manual annotation. This artifact mask guides the training of the AaDSA model, allowing it to bypass the adverse effects of artifact regions during model training. During inference, the AaDSA model can automatically generate artifact-free DSA images from single contrast images without any human intervention. Experimental results on a real head-and-neck DSA dataset show that our approach significantly outperforms state-of-the-art methods, highlighting its potential for clinical use.

Is the YES/NO classification accurate in screening scapular dyskinesis in asymptomatic individuals? - A novel validation study utilizing surface electromyography as a surrogate measure in identifying movement asymmetries

BACKGROUND

Scapular dyskinesis is a known risk factor for shoulder pain, making it important to screen for prevention. Physical therapists screen scapular dyskinesis by visually comparing asymmetries in scapular movement during overhead reach using the Scapular Dyskinesis Test Yes/No classification (Y/N). Although scapular kinematics has been used to quantify scapular dyskinesis, current measurement techniques are inaccurate. Optimal scapular muscle activity is crucial for normal shoulder function and is measured using surface electromyography (sEMG). Research suggests that impaired scapular muscles can lead to scapular dyskinesis. Despite kinematics being a poor reference standard, there is currently no validated method to identify movement asymmetries using muscle activity as an alternative. We utilized sEMG to establish Y/N's validity. We hypothesized that Y/N is a valid tool using sEMG as a viable surrogate measure for identifying scapular dyskinesis.

METHODS

We employed a known-groups (symmetrical vs. asymmetrical shoulders) validity design following the Standards for Reporting Diagnostic Accuracy Studies. Seventy-two asymptomatic subjects were evaluated using Y/N as the index test and sEMG as the reference standard. We created a criterion to assign the sEMG as the reference standard to establish the known groups. We calculated the sensitivity (Sn), specificity (Sp), positive and negative predictive values (PPV, NPV), likelihood ratios (LR+, LR-), and diagnostic odds ratio (DOR) using a 2 × 2 table analysis.

RESULTS

The diagnostic accuracy values were Sn = 0.56 (0.37-0.74), Sp = 0.36 (0.08-0.65), PPV = 0.68 (0.49-0.88), NPV = 0.25 (0.04-0.46), LR+ = 0.87 (0.50-1.53), and LR- = 1.22 (0.50-2.97).

CONCLUSION

The Y/N's diagnostic accuracy was poor against the sEMG, suggesting clinicians should rely less on Y/N to screen scapular dyskinesis in the asymptomatic population. Our study demonstrated that sEMG might be a suitable alternative as a reference standard in validating methods designed to screen movement asymmetries.

A systematic literature review of machine learning techniques for the detection of attention-deficit/hyperactivity disorder using MRI and/or EEG data

Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental condition common in teenagers across the globe. Neuroimaging and Machine Learning (ML) advancements have revolutionized its diagnosis and treatment approaches. Although, the researchers are continuously developing automated ADHD diagnostic tools, there is no reliable ML-based diagnostic system for clinicians. Thus, the study aims to systematically review ML and DL-based approaches for ADHD diagnosis, leveraging brain data from magnetic resonance imaging (MRI) and electroencephalogram (EEG) data. A methodical review for the period 2016 to 2022 is conducted by following the PRISMA guidelines. Four reputable repositories, namely PubMed, IEEE, ScienceDirect, and Springer are searched for the related literature on ADHD diagnosis using MRI/EEG data. 87 studies are selected after screening abstracts of the papers. We critically conducted an analysis of these studies by examining various aspects related to training ML/DL-models, including diverse datasets, hyperparameter tuning, overfitting, and interpretability. The quality and risk assessment is conducted using the QUADAS2 tool to determine the bias due to patient selection, index test, reference standard, and flow and timing. Our rigours analysis observed significant diversity in dataset acquisition and its size, feature extraction and selection techniques, validation strategies and classifier choices. Our findings emphasize the need for generalizability, transparency, interpretability, and reproducibility in future research. The challenges and potential solutions associated with integrating diagnostic models into clinical settings are also discussed. The identified research gaps will guide researchers in developing a reliable ADHD diagnostic system that addresses the associated challenges.

Total brain dose estimation in single-isocenter-multiple-targets (SIMT) radiosurgery via a novel deep neural network with spherical convolutions

BACKGROUND AND PURPOSE

Accurate prediction of normal brain dosimetric parameters is crucial for the quality control of single-isocenter multi-target (SIMT) stereotactic radiosurgery (SRS) treatment planning. Reliable dose estimation of normal brain tissue is one of the great indicators to evaluate plan quality and is used as a reference in clinics to improve potentially SIMT SRS treatment planning quality consistency. This study aimed to develop a spherical coordinate-defined deep learning model to predict the dose to a normal brain for SIMT SRS treatment planning.

METHODS

By encapsulating the human brain within a sphere, 3D volumetric data of planning target volume (PTVs) can be projected onto this geometry as a 2D spherical representation (in azimuthal and polar angles). A novel deep learning model spherical convolutional neural network (SCNN) was developed based on spherical convolution to predict brain dosimetric evaluators from spherical representation. Utilizing 106 SIMT cases, the model was trained to predict brain V50%, V60%, and V66.7%, corresponding to V10Gy and V12Gy, as key dosimetric indicators. The model prediction performance was evaluated using the coefficient of determination (R2), mean absolute error (MAE), and mean absolute percentage error (MAPE).

RESULTS

The SCNN accurately predicted normal brain dosimetric values from the modeled spherical PTV representation, with R2 scores of 0.92 ± 0.05/0.94 ± 0.10/0.93 ± 0.09 for V50%/V60%/V66.7%, respectively. MAEs values were 1.94 ± 1.61 cc/1.23 ± 0.98 cc/1.13 ± 0.99 cc, and MAPEs were 19.79 ± 20.36%/20.79 ± 21.07%/21.15 ± 22.24%, respectively.

CONCLUSIONS

The deep learning model provides treatment planners with accurate prediction of dose to normal brain, enabling improved consistency in treatment planning quality. This method can be extended to other brain-related analyses as an efficient data dimension reduction method.

Emerging trends in SERS-based veterinary drug detection: multifunctional substrates and intelligent data approaches

Veterinary drug residues in poultry and livestock products present persistent challenges to food safety, necessitating precise and efficient detection methods. Surface-enhanced Raman scattering (SERS) has been identified as a powerful tool for veterinary drug residue analysis due to its high sensitivity and specificity. However, the development of reliable SERS substrates and the interpretation of complex spectral data remain significant obstacles. This review summarizes the development process of SERS substrates, categorizing them into metal-based, rigid, and flexible substrates, and highlighting the emerging trend of multifunctional substrates. The diverse application scenarios and detection requirements for these substrates are also discussed, with a focus on their use in veterinary drug detection. Furthermore, the integration of deep learning techniques into SERS-based detection is explored, including substrate structure design optimization, optical property prediction, spectral preprocessing, and both qualitative and quantitative spectral analyses. Finally, key limitations are briefly outlined, such as challenges in selecting reporter molecules, data imbalance, and computational demands. Future trends and directions for improving SERS-based veterinary drug detection are proposed.

Development of heart-sparing VMAT radiotherapy technique incorporating heart substructures for advanced NSCLC patients

OBJECTIVE

To investigate the feasibility of active heart sparing (AHS) planning in patients with locally advanced and centrally located NSCLC receiving standard definitive radiotherapy (RT), while maintaining or improving appropriate lung, esophagus, and spinal cord constraints and planning target volume (PTV) coverage intent.

METHODS AND MATERIALS

A total of 27 patients with stage IIIA/B NSCLC treated with curative intent RT were selected for this analysis. All existing radiation plans were revised and 27 further new equivalent plans were calculated using AHS for the same cohort of patients. Primary end-point was feasibility of AHS using constraints for heart substructures. The secondary end point was to calculate the difference in terms of dosimetric parameters of heart substructures and principal OARs as well as PTV-coverage parameters within the current patient group.

RESULTS

AHS was feasible in the entire group of patients. An optimal coverage of the target volume was obtained and all mandatory constraints for OARs have been met. The median value of the mean heart dose (MHD) was 8.18 Gy and 6.71 Gy in the standard planning group and AHS-group, respectively (p = 0.000). Other heart parameters such as V5Gy (40.57% vs. 27.7%; p = 0.000) and V30Gy (5.39% vs. 3.86%; p = 0.000) were significantly worse in the standard planning group. The following relevant dosimetric parameters regarding heart substructures were found to be significantly worse in the standard planning group compared to the AHS-group: median dose to heart base (16.97 Gy vs. 6.37 Gy, p = 0.000), maximum dose (18.64 Gy vs. 6.05 Gy, p = 0.000) and V15Gy (11.11% vs. 0% p = 0.000) to LAD; mean dose; V5Gy (9.55% vs. 0.94%, p = 0.000) and V23Gy (0.00% vs. 0.00% maximum 45.68% vs. 6.57%, p = 0.002 to the left ventricle.

CONCLUSION

Our analysis showed an improvement of dosimetric parameters of the heart and heart substructures in patients affected by locally advanced and centrally located NSCLC treated with curative RT using AHS optimization. This approach could lead to a possible reduction of heart events and a prolonged survival. New clinical studies regarding RT in advanced NSCLC should include cardiologic evaluations and biomarkers as well as the contouring of cardiac substructures.

Exploring the significance of the frontal lobe for diagnosis of schizophrenia using explainable artificial intelligence and group level analysis

Schizophrenia (SZ) is a complex mental disorder characterized by a profound disruption in cognition and emotion, often resulting in a distorted perception of reality. Magnetic resonance imaging (MRI) is an essential tool for diagnosing SZ which helps to understand the organization of the brain. Functional MRI (fMRI) is a specialized imaging technique to measure and map brain activity by detecting changes in blood flow and oxygenation. The proposed paper correlates the results using an explainable deep learning approach to identify the significant regions of SZ patients using group-level analysis for both structural MRI (sMRI) and fMRI data. The study found that the heat maps for Grad-CAM show clear visualization in the frontal lobe for the classification of SZ and CN with a 97.33% accuracy. The group difference analysis reveals that sMRI data shows intense voxel activity in the right superior frontal gyrus of the frontal lobe in SZ patients. Also, the group difference between SZ and CN during n-back tasks of fMRI data indicates significant voxel activation in the frontal cortex of the frontal lobe. These findings suggest that the frontal lobe plays a crucial role in the diagnosis of SZ, aiding clinicians in planning the treatment.

Efficacy of a Supervised Exercise Program on Pain, Physical Function, and Quality of Life in Patients With Breast Cancer: Protocol for a Randomized Clinical Trial

BACKGROUND

Breast cancer is the second most common cancer in women worldwide. Treatments for this disease often result in side effects such as pain, fatigue, loss of muscle mass, and reduced quality of life. Physical exercise has been shown to effectively mitigate these side effects and improve the quality of life in patients with breast cancer.

OBJECTIVE

This randomized clinical trial aims to evaluate the efficacy of a 12-week supervised exercise program on pain, physical function, and quality of life in female patients with cancer.

METHODS

This randomized, double-blind clinical trial will recruit 325 participants, divided into an intervention group receiving the exercise program and a control group receiving standard care recommendations. Outcome measures, including pain (assessed via the Brief Pain Inventory), physical function (Disability of the Arm, Shoulder, and Hand Questionnaire), and quality of life (European Organization for Research and Treatment of Cancer QLQ-C30 and European Organization for Research and Treatment of Cancer QLQ-BR23), will be evaluated at baseline, immediately post intervention, and 12 weeks post intervention. Statistical analysis will involve repeated measures of ANOVA and MANOVA to determine the significance of the intervention's effects across time points.

RESULTS

Recruitment and data collection will commence in February of 2025, and data analysis is scheduled for completion at the end of 2025. No results are currently available.

CONCLUSIONS

Physical exercise is anticipated to play a significant role in alleviating pain, enhancing physical function, and improving the quality of life in female patients with cancer. This study will provide robust evidence to support the integration of supervised exercise into standard care protocols for this population.

TRIAL REGISTRATION

ClinicalTrials.gov NCT06618690; https://clinicaltrials.gov/ct2/show/NCT06618690.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

PRR1-10.2196/63891.

Stereotactic radiosurgery as neuromodulation of refractory angina: an initial case series

BACKGROUND

This intervention pilot case series assessed 40-Gy stereotactic radiosurgery (SRS) neuromodulation applied to the bilateral stellate ganglion (SG) as a bailout procedure for patients with refractory angina pectoris (RAP).

MATERIALS AND METHODS

The local institutional review board approved this feasibility study. In three patients with RAP, after repeated good response, symptoms were temporarily relieved after anaesthetic blockade of the left SG under ultrasound guidance. Radiosurgical neuromodulation with a dose of 40 Gy in one fraction was used for more permanent pain control. When RAP recurred after the initial SRS, right-sided procedures were considered after a confirmed positive response to right SG anesthetic block.

RESULTS

No acute or late radiation-related toxicities were observed. Two patients (67%) responded to bilateral SRS (follow-up: 60 and 48 months, respectively). From baseline to 24 months, their average prescribed nitrate package count decreased from 5.5 to 0 and remained low. Daily emergency nitrates declined from 20 to 30 to 1-2 applications, and walking distance improved from 10 to 20 m to 200-400 m and remained stable. Quality of life as measured with the EQ-5D and all domains of the Seattle Angina Questionnaire improved. The third patient received only unilateral SRS, had a temporary improvement for 6 months before a return to baseline, and died after 42 months of follow-up.

CONCLUSIONS

Bilateral radiosurgical neuromodulation at 40 Gy appears to be feasible, safe, and effective as a bailout procedure for RAP.

Sericin Electrodes with Self-Adhesive Properties for Biosignaling

The combination of green manufacturing approaches and bioinspired materials is growingly emerging in different scenarios, in particular in medicine, where widespread medical devices (MDs) as commercial electrodes for electrophysiology strongly increase the accumulation of solid waste after use. Electrocardiogram (ECG) electrodes exploit electrolytic gels to allow the high-quality recording of heart signals. Beyond their nonrecyclability/nonrecoverability, gel drying also affects the signal quality upon prolonged monitoring of biopotentials. Moreover, gel composition often causes skin reactions. This study aims to address the above limitation by presenting a composite based on the combination of silk sericin (SS) as a structural material, poly(vinyl alcohol) (PVA) as a robustness enhancer, and CaCl2 as a plasticizer. SS/PVA/CaCl2 formulations, optimized in terms of weight content (wt %) of single constituents, result in a biocompatible, biodegradable "green" material (free from potentially irritating cross-linking agents) that is, above all, self-adhesive on skin. The best formulation, i.e., SS(4 wt %)/PVA(4 wt %)/CaCl2(20 wt %), in terms of long-lasting skin adhesion (favored by calcium-ion coordination in the presence of environmental/skin humidity) and time-stability of electrode impedance, is used to assemble ECG electrodes showing quality trace recording over longer time scales (up to 6 h) than commercial electrodes. ECG recording is performed using customized electronics coupled to an app for data visualization.

Artificial intelligence-driven translational medicine: a machine learning framework for predicting disease outcomes and optimizing patient-centric care

BACKGROUND

Advancements in artificial intelligence (AI) and machine learning (ML) have revolutionized the medical field and transformed translational medicine. These technologies enable more accurate disease trajectory models while enhancing patient-centered care. However, challenges such as heterogeneous datasets, class imbalance, and scalability remain barriers to achieving optimal predictive performance.

METHODS

This study proposes a novel AI-based framework that integrates Gradient Boosting Machines (GBM) and Deep Neural Networks (DNN) to address these challenges. The framework was evaluated using two distinct datasets: MIMIC-IV, a critical care database containing clinical data of critically ill patients, and the UK Biobank, which comprises genetic, clinical, and lifestyle data from 500,000 participants. Key performance metrics, including Accuracy, Precision, Recall, F1-Score, and AUROC, were used to assess the framework against traditional and advanced ML models.

RESULTS

The proposed framework demonstrated superior performance compared to classical models such as Logistic Regression, Random Forest, Support Vector Machines (SVM), and Neural Networks. For example, on the UK Biobank dataset, the model achieved an AUROC of 0.96, significantly outperforming Neural Networks (0.92). The framework was also efficient, requiring only 32.4 s for training on MIMIC-IV, with low prediction latency, making it suitable for real-time applications.

CONCLUSIONS

The proposed AI-based framework effectively addresses critical challenges in translational medicine, offering superior predictive accuracy and efficiency. Its robust performance across diverse datasets highlights its potential for integration into real-time clinical decision support systems, facilitating personalized medicine and improving patient outcomes. Future research will focus on enhancing scalability and interpretability for broader clinical applications.

Determination of output correction factors in magnetic fields using two methods for two detectors at the central axis

Objective.Commissioning an MR-linac treatment planning system requires output correction factors,kB→,Qclin,Qmsrfclin,fmsr, for detectors to accurately measure the linac's output at various field sizes. In this study,kB→,Qclin,Qmsrfclin,fmsrwas determined at the central axis using two methods: one that combines the corrections for the influence of the magnetic field and the small field in a single factor (kB→,Qclin,Qmsrfclin,fmsr), and a second that isolates the magnetic field's influence, allowing the use of output correction factors without a magnetic field,kQclin,Qmsrfclin,fmsr, from literature for determiningkB→,Qclin,Qmsrfclin,fmsr.Approach.To determinekB→,Qclin,Qmsrfclin,fmsrand examine its behaviour across different photon energies and magnetic flux densitiesBin small fields, measurements with an ionization chamber (0.07 cm3sensitive volume) and a solid-state detector were carried out at an experimental facility for both approaches. Changes in absorbed dose to water with field size were determined via Monte Carlo simulations. To evaluate clinical applicability, additional measurements were conducted on a 1.5 T MR-linac.Main results.Both methods determined comparablekB→,Qclin,Qmsrfclin,fmsrresults. For field sizes >3 × 3 cm2,Branging from -1.5 to 1.5 T and photon energies of 6 and 8 MV, no change ofkQclin,Qmsrfclin,fmsras a function of the magnetic field was observed. Comparison with measurement results from the 1.5 T MR-linac confirm this. For ⩽3 × 3 cm2,kB→,Qclin,Qmsrfclin,fmsrdepends on photon energy andB. For 1.5 T and 6 MV,BreduceskQclin,Qmsrfclin,fmsrup to 3% for the ionization chamber and up to 7% for the solid-state detector.Significance.kB→,Qclin,Qmsrfclin,fmsrwere successfully determined for two detectors, enabling their use at a 1.5 T MR-linac. For field sizes of >3 × 3 cm2,kB→,Qclin,Qmsrfclin,fmsris one for most detectors suitable for small field dosimetry for all available perpendicular MR-linac systems, as confirmed in the literature. For these field sizes and detectors, the correction factor accounting for the dosimeter response change in the reference field due to the magnetic field,kB→,Qmsrfmsr, can be used for cross-calibration. Therefore, future research may only focus on small field sizes.

Diagnostic Performance of Artificial Intelligence-Based Methods for Tuberculosis Detection: Systematic Review

BACKGROUND

Tuberculosis (TB) remains a significant health concern, contributing to the highest mortality among infectious diseases worldwide. However, none of the various TB diagnostic tools introduced is deemed sufficient on its own for the diagnostic pathway, so various artificial intelligence (AI)-based methods have been developed to address this issue.

OBJECTIVE

We aimed to provide a comprehensive evaluation of AI-based algorithms for TB detection across various data modalities.

METHODS

Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) 2020 guidelines, we conducted a systematic review to synthesize current knowledge on this topic. Our search across 3 major databases (Scopus, PubMed, Association for Computing Machinery [ACM] Digital Library) yielded 1146 records, of which we included 152 (13.3%) studies in our analysis. QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies version 2) was performed for the risk-of-bias assessment of all included studies.

RESULTS

Radiographic biomarkers (n=129, 84.9%) and deep learning (DL; n=122, 80.3%) approaches were predominantly used, with convolutional neural networks (CNNs) using Visual Geometry Group (VGG)-16 (n=37, 24.3%), ResNet-50 (n=33, 21.7%), and DenseNet-121 (n=19, 12.5%) architectures being the most common DL approach. The majority of studies focused on model development (n=143, 94.1%) and used a single modality approach (n=141, 92.8%). AI methods demonstrated good performance in all studies: mean accuracy=91.93% (SD 8.10%, 95% CI 90.52%-93.33%; median 93.59%, IQR 88.33%-98.32%), mean area under the curve (AUC)=93.48% (SD 7.51%, 95% CI 91.90%-95.06%; median 95.28%, IQR 91%-99%), mean sensitivity=92.77% (SD 7.48%, 95% CI 91.38%-94.15%; median 94.05% IQR 89%-98.87%), and mean specificity=92.39% (SD 9.4%, 95% CI 90.30%-94.49%; median 95.38%, IQR 89.42%-99.19%). AI performance across different biomarker types showed mean accuracies of 92.45% (SD 7.83%), 89.03% (SD 8.49%), and 84.21% (SD 0%); mean AUCs of 94.47% (SD 7.32%), 88.45% (SD 8.33%), and 88.61% (SD 5.9%); mean sensitivities of 93.8% (SD 6.27%), 88.41% (SD 10.24%), and 93% (SD 0%); and mean specificities of 94.2% (SD 6.63%), 85.89% (SD 14.66%), and 95% (SD 0%) for radiographic, molecular/biochemical, and physiological types, respectively. AI performance across various reference standards showed mean accuracies of 91.44% (SD 7.3%), 93.16% (SD 6.44%), and 88.98% (SD 9.77%); mean AUCs of 90.95% (SD 7.58%), 94.89% (SD 5.18%), and 92.61% (SD 6.01%); mean sensitivities of 91.76% (SD 7.02%), 93.73% (SD 6.67%), and 91.34% (SD 7.71%); and mean specificities of 86.56% (SD 12.8%), 93.69% (SD 8.45%), and 92.7% (SD 6.54%) for bacteriological, human reader, and combined reference standards, respectively. The transfer learning (TL) approach showed increasing popularity (n=89, 58.6%). Notably, only 1 (0.7%) study conducted domain-shift analysis for TB detection.

CONCLUSIONS

Findings from this review underscore the considerable promise of AI-based methods in the realm of TB detection. Future research endeavors should prioritize conducting domain-shift analyses to better simulate real-world scenarios in TB detection.

TRIAL REGISTRATION

PROSPERO CRD42023453611; https://www.crd.york.ac.uk/PROSPERO/view/CRD42023453611.

On the feasibility of an online brain-computer interface-based neurofeedback game for enhancing attention and working memory in stroke and mild cognitive impairment patients

Background. Neurofeedback training (NFT) using Electroencephalogram-based Brain Computer Interface (EEG-BCI) is an emerging therapeutic tool for enhancing cognition.Methods. We developed an EEG-BCI-based NFT game for enhancing attention and working memory of stroke and Mild cognitive impairment (MCI) patients. The game involves a working memory task during which the players memorize locations of images in a matrix and refill them correctly using their attention levels. The proposed NFT was conducted across fifteen participants (6 Stroke, 7 MCI, and 2 non-patients). The effectiveness of the NFT was evaluated using the percentage of correctly filled matrix elements and EEG-based attention score. EEG varitions during working memory tasks were also investigated using EEG topographs and EEG-based indices.Results. The EEG-based attention score showed an enhancement ranging from 4.29-32.18% in the Stroke group from the first session to the third session, while in the MCI group, the improvement ranged from 4.32% to 48.25%. We observed significant differences in EEG band powers during working memory operation between the stroke and MCI groups.Significance. The proposed neurofeedback game operates based on attention and aims to improve multiple cognitive functions, including attention and working memory, in patients with stroke and MCI.Conclusions. The experimental results on the effect of NFT in patient groups demonstrated that the proposed neurofeedback game has the potential to enhance attention and memory skills in patients with neurological disorders. A large-scale study is needed in the future to prove the efficacy on a wider population.

A Thorough Review of the Clinical Applications of Artificial Intelligence in Lung Cancer

According to data from the World Health Organization (WHO), lung cancer is becoming a global epidemic. It is particularly high in the list of the leading causes of death not only in developed countries, but also worldwide; furthermore, it holds the leading place in terms of cancer-related mortality. Nevertheless, many breakthroughs have been made the last two decades regarding its management, with one of the most prominent being the implementation of artificial intelligence (AI) in various aspects of disease management. We included 473 papers in this thorough review, most of which have been published during the last 5-10 years, in order to describe these breakthroughs. In screening programs, AI is capable of not only detecting suspicious lung nodules in different imaging modalities-such as chest X-rays, computed tomography (CT), and positron emission tomography (PET) scans-but also discriminating between benign and malignant nodules as well, with success rates comparable to or even better than those of experienced radiologists. Furthermore, AI seems to be able to recognize biomarkers that appear in patients who may develop lung cancer, even years before this event. Moreover, it can also assist pathologists and cytologists in recognizing the type of lung tumor, as well as specific histologic or genetic markers that play a key role in treating the disease. Finally, in the treatment field, AI can guide in the development of personalized options for lung cancer patients, possibly improving their prognosis.

A deep learning model for radiological measurement of adolescent idiopathic scoliosis using biplanar radiographs

BACKGROUND

Accurate measurement of the spinal alignment parameters is crucial for diagnosing and evaluating adolescent idiopathic scoliosis (AIS). Manual measurement is subjective and time-consuming. The recently developed artificial intelligence models mainly focused on measuring the coronal Cobb angle (CA) and ignored the evaluation of the sagittal plane. We developed a deep-learning model that could automatically measure spinal alignment parameters in biplanar radiographs.

METHODS

In this study, our model adopted ResNet34 as the backbone network, mainly consisting of keypoint detection and CA measurement. A total of 600 biplane radiographs were collected from our hospital and randomly divided into train and test sets in a 3:1 ratio. Two senior spinal surgeons independently manually measured and analyzed spinal alignment and recorded the time taken. The reliabilities of automatic measurement were evaluated by comparing them with the gold standard, using mean absolute difference (MAD), intraclass correlation coefficient (ICC), simple linear regression, and Bland-Altman plots. The diagnosis performance of the model was evaluated through the receiver operating characteristic (ROC) curve and area under the curve (AUC). Severity classification and sagittal abnormalities classification were visualized using a confusion matrix.

RESULTS

Our AI model achieved the MAD of coronal and sagittal angle errors was 2.15° and 2.72°, and ICC was 0.985, 0.927. The simple linear regression showed a strong correction between all parameters and the gold standard (p < 0.001, r2 ≥ 0.686), the Bland-Altman plots showed that the mean difference of the model was within 2° and the automatic measurement time was 9.1 s. Our model demonstrated excellent diagnostic performance, with an accuracy of 97.2%, a sensitivity of 96.8%, a specificity of 97.6%, and an AUC of 0.972 (0.940-1.000).For severity classification, the overall accuracy was 94.5%. All accuracy of sagittal abnormalities classification was greater than 91.8%.

CONCLUSIONS

This deep learning model can accurately and automatically measure spinal alignment parameters with reliable results, significantly reducing diagnostic time, and might provide the potential to assist clinicians.

Autonomic correlates of osteopathic manipulative treatment on facial functional mapping: an innovative approach based on thermal imaging

Osteopathic manipulative treatment (OMT) has shown efficacy in various clinical conditions and age groups. Understanding its neurobiological, particularly autonomic, mechanisms of action remain limited. Preliminary studies suggested a parasympathetic effect of OMT, evidenced by heart-rate-variability analysis. A cross-over RCT on healthy adults was conducted to compare OMT with sham therapy. Thirty-seven participants underwent two sessions (OMT and sham), comprising baseline, tactile treatment, and post-touch. Novel thermal imaging data analyses in combination with seed correlation analyses (SCA) were employed to explore the OMT effects on autonomic parameters. Particularly, the sham group exhibited an elevated warming effect on the cheeks, nose, and chin. Inversely, for the OMT group a conspicuous cooling trend in the nose, but not in the cheeks and chin was observed. Considering SCA maps, the intensity of the correlation for nose tip, glabella and GSR seeds showed higher values in the OMT compared to the sham group. The comparative analysis of thermal maps and SCA results represents a significant advancement in our understanding of the physiological mechanisms underlying OMT's effects on autonomic functions. By elucidating specific patterns of temperature change, correlation intensity and specific clusters, this research provides valuable insights for optimizing clinical practice and refining theoretical models of manual therapy.

ESTRO guidelines for developing questionnaires in survey-based radiation oncology research

Survey-based research is increasingly used in radiation oncology, yet survey-based research methodology is often unfamiliar in the field. This guideline offers comprehensive instructions for conducting survey-based research in radiation oncology, covering critical aspects such as survey design, validation, dissemination, analysis, and reporting. Tailored to professionals, it emphasizes the importance of methodological rigour to ensure reliable and actionable data collection. Dissemination strategies are highlighted to maximize response rates and enhance data completeness across diverse clinical, research and industrial settings. Rigorous analysis techniques are discussed to uncover insights that optimize operational efficiencies and inform evidence-based practices. Transparent reporting is underscored as crucial for enhancing the credibility and applicability of findings. This guideline aims to be a practical resource for enhancing research standards in survey-based research for researchers and practitioners in radiation oncology.

Photon-Counting Detector CT Applications in Musculoskeletal Radiology

ABSTRACT

Photon-counting detectors (PCDs) have emerged as one of the most influential technical developments for medical imaging in recent memory. Surpassing conventional systems with energy-integrating detector technology in many aspects, PCD-CT scanners provide superior spatial resolution and dose efficiency for all radiological subspecialities. Demanding detailed display of trabecular microarchitecture and extensive anatomical coverage frequently within the same scan, musculoskeletal (MSK) imaging in particular can be a beneficiary of PCD-CT's remarkable performance. Since PCD-CT provides users with a plethora of customization options for both image acquisition and reconstruction, however, MSK radiologists need to be familiar with the scanner to unlock its full potential. From filter-based spectral shaping for artifact reduction over full field-of-view ultra-high-resolution scans to postprocessing of single- or dual-source multienergy data, almost every imaging task can be met with an optimized approach in PCD-CT. The objectives of this review were to give an overview of the most promising applications of PCD-CT in MSK imaging to date, to state current limitations, and to highlight directions for future research and developments.

Federated learning for enhanced dose-volume parameter prediction with decentralized data

BACKGROUND

The widespread adoption of knowledge-based planning in radiation oncology clinics is hindered by the lack of data and the difficulty associated with sharing medical data.

PURPOSE

This study aims to assess the feasibility of mitigating this challenge through federated learning (FL): a centralized model trained with distributed datasets, while keeping data localized and private.

METHODS

This concept was tested using 273 prostate 45 Gy plans. The cases were split into a training set with 220 cases and a validation set with 53 cases. The training set was further separated into 10 subsets to simulate treatment plans from different clinics. A gradient-boosting model was used to predict bladder and rectum V30Gy, V35Gy, and V40Gy. The Federated Averaging algorithm was employed to aggregate the individual model weights from distributed datasets. Grid search with five-fold in-training-set cross-validation was implemented to tune model hyperparameters. Additionally, we evaluated the robustness of the FL approach by varying the distribution of the training set data in several scenarios, including different number of sites and imbalanced data across sites.

RESULTS

The mean absolute error (MAE) for the FL model (4.7% ± 2.9%) is significantly lower than individual models trained separately (6.5% ± 4.9%, p < 0.001) and similar to a traditional centralized model (4.4% ± 2.8%, p = 0.14). The federated model is robust to the number of subsets, showing MAE of 4.7% ± 3.2%, 4.8% ± 3.1%, 4.8% ± 2.9%, 4.5% ± 2.8%, 4.9% ± 3.3%, and 4.8% ± 3.1% for 5, 10, 15, 20, 25, and 30 subsets, respectively. For the two imbalanced datasets, the FL model achieves MAEs of 4.5% ± 2.9% and 5.6% ± 4.0%, non-inferior to the balanced data model. For all bladder and rectum metrics, the FL model significantly outperforms 36.7% of individual models.

CONCLUSIONS

This study demonstrates the potential advantages of implementing a federated model over training individual models: the proposed FL approach achieves similar prediction accuracy as a conventional model without requiring centralized data storage. Even when local models struggle to produce accurate predictions due to data scarcity, the federated model consistently maintains high performance.

PADS-Net: GAN-based radiomics using multi-task network of denoising and segmentation for ultrasonic diagnosis of Parkinson disease

Parkinson disease (PD) is a prevalent neurodegenerative disorder, and its accurate diagnosis is crucial for timely intervention. We propose the PArkinson disease Denoising and Segmentation Network (PADS-Net), to simultaneously denoise and segment transcranial ultrasound images of midbrain for accurate PD diagnosis. The PADS-Net is built upon generative adversarial networks and incorporates a multi-task deep learning framework aimed at optimizing the tasks of denoising and segmentation for ultrasound images. A composite loss function including the mean absolute error, the mean squared error and the Dice loss, is adopted in the PADS-Net to effectively capture image details. The PADS-Net also integrates radiomics techniques for PD diagnosis by exploiting high-throughput features from ultrasound images. A four-branch ensemble diagnostic model is designed by utilizing two "wings" of the butterfly-shaped midbrain regions on both ipsilateral and contralateral images to enhance the accuracy of PD diagnosis. Experimental results demonstrate that the PADS-Net not only reduced speckle noise, achieving the edge-to-noise ratio of 16.90, but also attained a Dice coefficient of 0.91 for midbrain segmentation. The PADS-Net finally achieved an area under the receiver operating characteristic curve as high as 0.87 for diagnosis of PD. Our PADS-Net excels in transcranial ultrasound image denoising and segmentation and offers a potential clinical solution to accurate PD assessment.

Rectal Gas-Induced Dose Changes in Carbon Ion Radiation Therapy for Prostate Cancer: An In Silico Study

Purpose

This study aims to determine dosimetric influence of rectal gas in carbon ion radiation therapy (CIRT) for prostate cancer and to establish a procedure for removal rectal gas in clinical scenarios.

Materials and methods

We analyzed 18 prostate cancer cases with bulky rectal gas. The dose distribution was recalculated on computed tomography (CT) with bulky rectal gas (gasCT) after creating the initial plan on a CT without bulky rectal gas, and the doses were transformed using a displacement vector field. This created a dose distribution simulation irradiated with the residual rectal gas. Among 12 fractions (fx) for prostate cancer CIRT, different residual rectal gas fx were used to develop 12 dose distributions, each of which was compared with that in the initial plan. Clinical target volume (Dmin, D99.5%), rectum, and rectal wall (V95%, V80%) parameters were assessed. We investigated the indicators associated with these dose changes using digital reconstruction radiograph (DRR) images.

Results

The dosimetric changes in the clinical target volume were not significantly different from that in the initial treatment plan for both Dmin and D99.5%. Compared to the initial plan, the dose-volume histogram parameters showed changes exceeding 1 cm3 when residual rectal gas was present in the following number of fractions: 8 fx for V95% rectum, 5 fx for V80% rectum, 10 fx for V95% rectal wall, and 11 fx for V80% rectal wall. Changes in rectal and rectal wall parameters were highly correlated with the extent of rectal gas assessed on DRR images.

Conclusion

Rectal gas removal may not be necessary up to 4 fx. Moreover, indicators related to dose changes based on DRR images were highly correlated with dose changes, revealing the possibilities of estimating dose changes due to rectal gas from kV-x-ray images and using gas effect evaluation during CIRT irradiation.

The clinical application of in vivo dosimetry for gynaecological brachytherapy: A scoping review

Brachytherapy is a key treatment for gynaecological malignancies, delivering high doses to the tumour volume whilst sparing nearby normal tissues due to its steep dose gradient. Accuracy is imperative as small shifts can lead to clinically significant under- or over-dosing of the target volume or organs at risk (OARs), respectively. Independent verification of dose delivered during brachytherapy is not routinely performed but it is important to identify gross errors and define action thresholds to guide inter-fraction treatment decisions. In vivo dosimetry (IVD) is one strategy for improving accuracy and identifying potential errors. Despite promising phantom work, clinical application of IVD is lacking. A literature search was performed using Medline and EMBASE without date limits and based on the PICO framework to evaluate the clinical application of IVD in gynaecological brachytherapy. After screening of titles and abstracts, full text papers were reviewed and 28 studies were identified. Several dosimeters were utilised and measurements were typically taken from the rectum, bladder, vagina and within interstitial catheters. Significant differences between calculated and measured dose were attributed to geometric shifts. The studies reviewed demonstrated the feasibility of IVD in brachytherapy for dose verification but further work is required before IVD can be used to optimise treatment. The purpose of this scoping review is to investigate the clinical application of IVD in gynaecological brachytherapy, understand its challenges and identify the steps required to facilitate integration into everyday clinical practice.

Schizophrenia detection using distributed activation function-based statistical attentional bidirectional-long short-term memory

Schizophrenia detection involves identifying the schizophrenia by analyzing specific patterns in Electroencephalogram (EEG) signals, which reflect brain activity associated with symptoms, like hallucinations and cognitive impairments. Existing models face challenges due to the complex and variable nature of EEG data, which may struggle to accurately capture critical temporal dependencies and relevant features. Traditional approaches often lack adaptability, limiting their ability to differentiate schizophrenia patterns from other brain activities. Hence, a Distributed Activation function-based statistical Attention Bi-LSTM (DA-SA-BiLSTM) is proposed for schizophrenia detection, which enhances the precision and interpretability of EEG signal analysis. This model effectively manages the temporal dependencies for the detection as it incorporates past and future data context to improve decision-making. By dynamically weighting features based on their relevance, the model emphasizes critical segments and reduces noise, increasing predictive accuracy. Using different activation functions in various layers, the DA-AB-LSTM is allowed to adapt to specific characteristics of the EEG data, strengthening its flexibility and pattern recognition abilities. Furthermore, this model refines relationships between features, facilitating precise class probability distribution for schizophrenia classification. In particular, the DA-SA-BiLSTM model outperforms the existing models with 95.9 % accuracy, the lowest mean square error (MSE) of 5.86, 95.84 % sensitivity, and 95.97 % specificity.

Super-resolution deep-learning reconstruction with 1024 matrix improves CT image quality for pancreatic ductal adenocarcinoma assessment

OBJECTIVES

To evaluate the efficiency of super-resolution deep-learning reconstruction (SR-DLR) optimized for helical body imaging in assessing pancreatic ductal adenocarcinoma (PDAC) using normal-resolution (NR) CT scanner.

METHODS

Fifty patients with PDAC who underwent multiphase pancreas CT on a 320-row NR scanner were retrospectively analyzed. Images were reconstructed using hybrid iterative reconstruction (HIR), normal-resolution deep-learning reconstruction (NR-DLR), and SR-DLR at a 0.5-mm slice thickness. The matrix size was 512 × 512 for HIR and NR-DLR, and 1024 × 1024 for SR-DLR. Image noise and contrast-to-noise ratio (CNR) of pancreas, superior mesenteric artery, portal vein, and PDAC were quantified. Noise power spectrum (NPS) in the liver and edge rise slope (ERS) at the pancreas, artery, and vein were used to quantify noise properties and edge sharpness. Subjective evaluations included rankings of image sharpness, noise magnitude, texture fineness, and delineation of PDAC, pancreas margin, pancreatic duct, peripancreatic vessels, and hepatic lesions (1 = worst; 3 = best among three image series). Overall diagnostic quality was rated on a 5-point scale (1 = undiagnostic, 5 = excellent).

RESULTS

SR-DLR showed significantly lower image noise and higher CNR than HIR and NR-DLR (all, p < 0.001). NPS analysis revealed no significant difference in average spatial frequency between SR-DLR and NR-DLR (p = 0.770), both being higher than HIR (both, p < 0.001). ERS values of all structures were highest with SR-DLR (p < 0.001). SR-DLR received the highest subjective scores for all criteria, with significant differences from HIR and NR-DLR (all, p < 0.001).

CONCLUSION

SR-DLR improved both subjective and objective image quality, enhancing the delineation of all structures relevant to PDAC assessment using NR CT scanner.

Working thresholds for in-vivo dosimetry in EPIGray based on a clinical, anatomically-stratified study

PURPOSE

To obtain tolerance levels for working with the EPID-based EPIgray in vivo dosimetry system.

METHODS

Dose differences between planned and delivered treatments in various anatomical areas, including the gastro-intestinal, urological, rectum and anal canal, gynecological, breast, head and neck, and lung regions, were analyzed across 5,791 fractions. Whether or not the dose differences at each location are symmetrical with respect to zero and adhere to a normal distribution is then checked. Linear regression analysis was applied to check for temporal drift in lung and head and neck treatments. A water equivalent phantom and another with a water-polystyrene interface is used to estimate the dose difference intrinsic to the measurement system. Furthermore, appropriate dose distribution in two treatments is verified using radiochoromic film.

RESULTS

Normal distribution was not observed in any region, and only two showed symmetry around zero. The mean dose differences were: (0.33 ± 6.32) % for the gastro-intestinal system, (-1.31 ± 3.16) % for the gynaecological area, (0.79 ± 4.55) % for VMAT-breast, (3.48 ± 4.00) % for 3DCRT-breast, (0.70 ± 3.20) % for head and neck, (5.63 ± 5.48)% for lung, (-1.36 ± 2.98) % for rectum and anal canal, and (0.13 ± 3.53) % for the urological system.

CONCLUSION

EPIgray should support tolerance levels asymmetric with respect to zero, given the positive deviation observed in mean dose for lung, breast, and head and neck regions. Additionally, the system's ability to detect dose variations during treatment could help identify changes in tumor volume.

Comprehensive image quality comparison of conventional and new flat panel detectors under bedside chest radiography beam conditions

Recently, a novel wireless flat-panel detector with auto-exposure control has become available. This study aimed to elucidate the potential advantages of the new detector over conventional detectors through a comprehensive analysis of the physical image quality characteristics. Measurements were conducted on two models: new (720C) and conventional (710C) versions; this assessment was performed by assuming the beam quality for bedside chest radiography, utilizing a portable device for X-ray exposure. The detective quantum efficiency (DQE) was computed based on the presampled modulation transfer function (MTF) and normalized noise power spectrum. The validity of the DQE results was verified through the visualization of the analog blurring components and a detailed analysis of the noise components. The spatial frequency at which the presampled MTF value reached 10% was 5.2 cycles/mm for 720C and 3.9 cycles/mm for 710C. The full width at half-maximum of the spatial spreading of analog components was estimated at 0.09 mm for 720C and 0.14 mm for 710C by the visualization. Regarding the DQE, 720C was superior under low-dose conditions despite no significant differences being observed under high-dose conditions. The new detector demonstrated superior resolution characteristics compared with the conventional detector and an improvement in the DQE under low-dose conditions. However, similar to the conventional detector, a significant dose dependence caused by a structural factor was confirmed for the DQE. These results suggest the existence of an appropriate dose range for maximizing detector performance and provide insights crucial for optimization tasks in the X-ray imaging.

Effect of Ethanol on Brain Electrical Tissue Conductivity in Social Drinkers

BACKGROUND

How the biophysics of electrical conductivity measures relate to brain activity is poorly understood. The sedative, ethanol, reduces metabolic activity but its impact on brain electrical conductivity is unknown.

PURPOSE

To investigate whether ethanol reduces brain electrical tissue conductivity.

STUDY TYPE

Prospective.

SUBJECTS

Fifty-two healthy volunteers (aged 18-37 years, 22 females, 30 males).

FIELD STRENGTH/SEQUENCE

3 T, T1-weighted, multi-shot, turbo-field echo (TFE); 3D balanced fast-field echo (bFFE).

ASSESSMENT

Brain gray and white matter tissue conductivity measured with phase-based magnetic resonance electrical properties tomography (MREPT) compared before and 20 minutes after ethanol consumption (0.7 g/kg body weight). Differential conductivity whole brain maps were generated for three subgroups: those with strong ( ∆ σ max  > 0.1 S/m; N = 33), weak (0.02 S/m ≤  ∆ σ max  ≤ 0.1 S/m; N = 9) conductivity decrease, and no significant response ( ∆ σ max  < 0.02 S/m, N = 10). Maps were compared in the strong response group where breath alcohol rose between scans, vs. those where it fell.

STATISTICAL TESTS

Average breath alcohol levels were compared to the differential conductivity maps using linear regression. T-maps were generated (threshold P < 0.05 and P < 0.001; minimum cluster 48 mm3). Differential conductivity maps were compared with ANOVA.

RESULTS

Whole-group analysis showed decreased conductivity that did not survive statistical thresholding. Strong responders (N = 33) showed a consistent pattern of significantly decreased conductivity ( ∆ σ max  > 0.1 S/m) in frontal/occipital and cerebellar white matter. The weak response group (N = 9) showed a similar pattern of conductivity decrease (0.02 S/m ≤  ∆ σ max  ≤ 0.1 S/m). There was no significant relationship with breath alcohol levels, alcohol use, age, ethnicity, or sex. The strong responders' regional response was different between ascending (N = 12) or descending (N = 20) alcohol during the scan.

DATA CONCLUSION

Ethanol reduces brain tissue conductivity in a participant-dependent and spatially dependent fashion.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 2.

Technique considerations for implementing volumetric-modulated arc therapy for total body irradiation within an Australian tertiary institution

Over the past decade, our institution delivered conventional total body irradiation (TBI) using Elekta's Monaco treatment planning system (TPS) with an extended SSD field arrangement and 18 megavoltage (MV) energy lateral fields. In 2020, there was a transition to the Eclipse™ treatment planning system and Truebeam® linear accelerators with 6 MV and 10 MV energies. These changes meant that essential components of the existing technique were unavailable for clinical use and a new approach to the institution technique was required to ensure continuation of service. The aim was to implement a volumetric-modulated arc therapy (VMAT) TBI technique using existing infrastructure, the new planning system and treatment hardware to continue providing a TBI service for patients of all ages, including those under general anaesthetic (GA). A multidisciplinary team within the institution was created to evaluate existing literature and to implement a VMAT TBI technique that was feasible within our institution. This article will discuss the resultant technique, the practicalities faced and the radiation therapy pathway within our institution.

Practical X-ray gastric cancer diagnostic support using refined stochastic data augmentation and hard boundary box training

Endoscopy is widely used to diagnose gastric cancer and has a high diagnostic performance, but it must be performed by a physician, which limits the number of people who can be diagnosed. In contrast, gastric X-rays can be taken by radiographers, thus allowing a much larger number of patients to undergo imaging. However, the diagnosis of X-ray images relies heavily on the expertise and experience of physicians, and few machine learning methods have been developed to assist in this process. We propose a novel and practical gastric cancer diagnostic support system for gastric X-ray images that will enable more people to be screened. The system is based on a general deep learning-based object detection model and incorporates two novel techniques: refined probabilistic stomach image augmentation (R-sGAIA) and hard boundary box training (HBBT). R-sGAIA enhances the probabilistic gastric fold region and provides more learning patterns for cancer detection models. HBBT is an efficient training method that improves model performance by allowing the use of unannotated negative (i.e., healthy control) samples, which are typically unusable in conventional detection models. The proposed system achieved a sensitivity (SE) for gastric cancer of 90.2%, higher than that of an expert (85.5%). Under these conditions, two out of five candidate boxes identified by the system were cancerous (precision = 42.5%), with an image processing speed of 0.51 s per image. The system also outperformed methods using the same object detection model and state-of-the-art data augmentation by showing a 5.9-point improvement in the F1 score. In summary, this system efficiently identifies areas for radiologists to examine within a practical time frame, thus significantly reducing their workload.

AutoCOR: Autonomous condylar offset ratio calculator for post-operative total knee arthroplasty radiographs

BACKGROUND

This study aims to automate the measurement process of posterior condylar offset ratio (PCOR) and anterior condylar offset ratio (ACOR) to improve the Total Knee Arthroplasty (TKA) evaluation. Accurate calculation of PCOR and ACOR, performed manually by orthopedic surgeons, is crucial for assessing postoperative range of motion and implant positioning. Manual measurements, however, are time-consuming, prone to human error, and subject to variability. Automating this process could improve precision in clinical practice.

METHODS

We developed AutoCOR, a software system that autonomously calculates PCOR and ACOR by utilizing built-in function, employing k-means clustering, from the OpenCV library for image segmentation. The software detects key anatomical landmarks on true postoperative lateral radiographs. The definitions of PCOR and ACOR are PCO (posterior condylar offset) divided by femoral diameter, and ACOR is defined as ACO (anterior condylar offset) divided by femoral diameter, respectively. We tested the algorithm on 50 postoperative lateral radiographs of 32 patients from the Istanbul Kosuyolu Medipol Hospital, which included data from. The assessment process included calculating the mean, standard deviation and plotting the Bland-Altman plots, comparing AutoCOR's results against ground truth values.

RESULTS

The mean PCOR was 0.984 (SD 0.235) for AutoCOR and 0.972 (SD 0.164) for ground truth values, showing a strong correlation (Pearson r = 0.845, p < 0.0001). The mean ACOR was 0.107 (SD 0.092) for AutoCOR and 0.107 (SD 0.070) for ground truth values, with moderate correlation (Spearman's rs = 0.519, p = 0.0001).

CONCLUSION

AutoCOR provides accurate measurements and shows potential to reduce variability in TKA evaluation, improving precision in clinical practice.

Deep Learning-Based Artificial Intelligence Can Differentiate Treatment-Resistant and Responsive Depression Cases with High Accuracy

Background: Although there are many treatment options available for depression, a large portion of patients with depression are diagnosed with treatment-resistant depression (TRD), which is characterized by an inadequate response to antidepressant treatment. Identifying the TRD population is crucial in terms of saving time and resources in depression treatment. Recently several studies employed various methods on EEG datasets for automatic depression detection or treatment outcome prediction. However, no previous study has used the deep learning (DL) approach and EEG signals for detecting treatment resistance. Method: 77 patients with TRD, 43 patients with non-TRD, and 40 healthy controls were compared using GoogleNet convolutional neural network and DL on EEG data. Additionally, Class Activation Maps (CAMs) acquired from the TRD and non-TRD groups were used to obtain distinctive regions for classification. Results: GoogleNet classified the healthy controls and non-TRD group with 88.43%, the healthy controls and TRD subjects with 89.73%, and the TRD and non-TRD group with 90.05% accuracy. The external validation accuracy for the TRD-non-TRD classification was 73.33%. Finally, the CAM analysis revealed that the TRD group contained dominant features in class detection of deep learning architecture in almost all electrodes. Limitations: Our study is limited by the moderate sample size of clinical groups and the retrospective nature of the study. Conclusion: These findings suggest that EEG-based deep learning can be used to classify treatment resistance in depression and may in the future prove to be a useful tool in psychiatry practice to identify patients who need more vigorous intervention.

Diabetic retinopathy detection via deep learning based dual features integrated classification model

BackgroundThe primary recognition of diabetic retinopathy (DR) is a pivotal requirement to prevent blindness and vision impairment. This deadly condition is identified by highly qualified professionals by examining colored retinal images.ObjectiveThe physical diagnostics for this condition was time-consuming and prone to fault. The development of computer-vision based intelligent systems has develop a main research area to effectually diagnosis the pathologies from an image.MethodsIn this research, a novel Deep learning based Dual Features Integrated classification (DD-FIC) framework is designed to detect the DR from a color retinal image. Initially, the fundus images are denoised by Wavelet integrated Retinex (WIR) algorithm to remove the noise artifacts which provide high contrast image. This DD-FIC model contains two phases of feature extraction module to evaluation of several retinal areas. Initially, global features of the fundus image are retrieved by the assist of attention fused efficient model, whereas the attention module dynamically highlights the important features. Afterwards, the segmented retinal vessels data is converted into features for learning the local features.ResultsFinally, the collective of features is processed into the Random Forest based feature selection model for the optimal prediction with five different classes using multi-class support vector machine (MCSVM). The efficacy of the proposed DD-FIC framework is estimated by Kaggle dataset with the detection accuracy of 98.6%.Conclusions: The proposed framework rises the accuracy of 1.54%, 3.65%, 13.79% and 6.28% for Multi-channel CNN, CNN, VGG NiN and Shallow CNN respectively.

A novel approach to enhancing biomedical signal recognition via hybrid high-order information bottleneck driven spiking neural networks

Biomedical signals, encapsulating vital physiological information, are pivotal in elucidating human traits and conditions, serving as a cornerstone for advancing human-machine interfaces. Nonetheless, the fidelity of biomedical signal interpretation is frequently compromised by pervasive noise sources such as skin, motion, and equipment interference, posing formidable challenges to precision recognition tasks. Concurrently, the burgeoning adoption of intelligent wearable devices illuminates a societal shift towards enhancing life and work through technological integration. This surge in popularity underscores the imperative for efficient, noise-resilient biomedical signal recognition methodologies, a quest that is both challenging and profoundly impactful. This study proposes a novel approach to enhancing biomedical signal recognition. The proposed approach employs a hierarchical information bottleneck mechanism within SNNs, quantifying the mutual information in different orders based on the depth of information flow in the network. Subsequently, these mutual information, together with the network's output and category labels, are restructured based on information theory principles to form the loss function used for training. A series of theoretical analyses and substantial experimental results have shown that this method can effectively compress noise in the data, and on the premise of low computational cost, it can also significantly outperform its vanilla counterpart in terms of classification performance.

A machine learning approach for diagnostic and prognostic predictions, key risk factors and interactions

Machine learning (ML) has the potential to revolutionize healthcare, allowing healthcare providers to improve patient-care planning, resource planning and utilization. Furthermore, identifying key-risk-factors and interaction-effects can help service-providers and decision-makers to institute better policies and procedures. This study used COVID-19 electronic health record (EHR) data to predict five crucial outcomes: positive-test, ventilation, death, hospitalization days, and ICU days. Our models achieved high accuracy and precision, with AUC values of 91.6%, 99.1%, and 97.5% for the first three outcomes, and MAE of 0.752 and 0.257 days for the last two outcomes. We also identified interaction effects, such as high bicarbonate in arterial blood being associated with longer hospitalization in middleaged patients. Our models are embedded in a prototype of an online decision support tool that can be used by healthcare providers to make more informed decisions.

Shear-Wave Elastography Improves Diagnostic Accuracy in Chronic Kidney Disease Compared to Conventional Ultrasound

PURPOSE

Non-invasive tests are increasingly demanded for diagnosing and prognostication of chronic kidney disease (CKD). Shear-wave elastography (SWE), an emerging technique for measuring tissue stiffness, shows promise for distinguishing between individuals with different stages of renal fibrosis. This study aimed to compare the diagnostic accuracy of two-dimensional SWE (2D-SWE) and conventional ultrasound for detecting CKD, employing renal biopsy as the gold standard.

METHODS

From May 2020 to October 2023, this prospective study included 30 healthy volunteers and 169 patients with CKD who had undergone 2D-SWE and conventional ultrasound of both kidneys. Cortical and medullary stiffness, cortical pixel intensity, renal length, parenchymal and cortical thickness, interlobar artery peak systolic velocity, end-diastolic velocity (EDV), and resistive index were measured. The diagnostic accuracy of 2D-SWE and conventional ultrasound was compared using the receiver operating characteristic curve (ROC) and Delong test.

RESULTS

For diagnosing CKD, the area under the ROC (AUC) of cortical stiffness (0.96 [95% CI, 0.93, 0.99]) was significantly higher than that of all conventional ultrasound parameters, including EDV (0.78 [95% CI, 0.71, 0.86]) and cortical thickness (0.74 [95% CI, 0.67, 0.80]). The sensitivity of cortical stiffness (91%) was significantly higher than that of EDV (68%) and cortical thickness (53%). No significant difference was found in the specificity of cortical stiffness (96%) compared to that of EDV (79%) and cortical thickness (100%).

CONCLUSION

Two-dimensional SWE showed higher diagnostic accuracy than that of conventional ultrasound for detecting CKD.

Transforming Brain Tumor Care: The Global Impact of Radiosurgery in Multidisciplinary Treatment Over Two Decades

BACKGROUND

Stereotactic radiosurgery, a minimally invasive treatment delivering high doses of radiation to a well-defined target, has transformed interdisciplinary treatment paradigms since its inception. This study chronicles its adoption and evolution for brain cancer and tumors globally.

METHODS

A systematic literature review of SRS-focused articles from 2000 to 2023 was conducted. Literature impact was evaluated using citation counts and relative citation ratio scores. Extracted data were dichotomized between US and international publications.

RESULTS

Out of 5424 articles eligible, 538 met inclusion criteria reporting on 120,756 patients treated with SRS for brain cancer and tumors since 2000. Over time, publication rates grew significantly (p = 0.0016), with 56% of principal investigators based in the United States. Clinical articles accounted for 87% of the publications, with the remainder focused on technological advances. Relative to international studies, US publications had larger median samples (74 vs. 58, p = 0.012), higher median citations (30 vs. 19, p < 0.0001) and higher relative citation ratio scores (1.67 vs. 1.2, p < 0.00001). Gamma Knife and LINAC had roughly equal representation in US and international publications. Neurosurgery specialists authored more Gamma Knife-based articles, and radiation oncology specialists authored more LINAC-based papers (p < 0.0001). The most treated tumors were metastases (58%), skull base tumors (35%), and gliomas (7%). Radiographic control was achieved in 82% of metastatic tumor cases, with a 12% median complication rate.

CONCLUSIONS

SRS has been widely adopted both nationally and globally and continues to be a growing field. This study corroborates the clinical efficacy of SRS and reinforces its critical role in the multidisciplinary treatment of patients with brain tumors and cancer.

Quality assessment of automatically planned o-ring linac SBRT plans for pelvic lymph node and lung metastases, evaluating the optimal minimum target size

The purpose of this study is to assess the influence of Planning Target Volume (PTV) on the quality of automatic planned O-Ring Halcyon linac stereotactic body radiation therapy (SBRT) plans of pelvic lymph nodes (LN) and lung metastases and to evaluate an absolute PTV volume threshold as a plan quality prediction criterion. A total of 21 pelvic LN and 18 lung clinical treatment plans were replanned for Halcyon with unattended autoplanning. The prescription dose range was 26-40 Gy for LN and between 39-54 Gy for the lung in the mean 3 fractions. The mean/median PTV was 4.0/ 3.6 cm3 for LN and 4.9/ 4.3 cm3 for the lung. The criteria for the plan quality evaluation consisted of using dose metrics for conformity, spillage, and coverage and dose limits on healthy tissue assessment. A statistical study was performed based on systematic Mann-Whitney U test and cluster analysis to evaluate a PTV volume predictor threshold of plan quality. 95% (n = 20) LN and 100% (n = 18) lung plans met all tolerance criteria. For both cohorts of plans, a PTV threshold was determined, indicating a reduction of particular dose indices when below this threshold. Low risk of toxicity in healthy tissues was predicted. A PTV threshold of 4.0 cm3 was estimated as quality criteria in both cohorts of plans. The results of our study demonstrated the promising performance of Halcyon for pelvic and lung SBRT for small tumors, although plan-specific QA is required to verify machine performance during plan delivery.

A Proof-of-Concept Solution for Co-locating 2D Histology Images in 3D for Histology-to-CT and MR Image Registration: Closing the Loop for Bone Sarcoma Treatment Planning

This work presents a proof-of-concept solution designed to facilitate more accurate radiographic feature characterisation in pre-surgical CT/MR volumes. The solution involves 3D co-location of 2D digital histology slides within ex-vivo, tumour tissue CT volumes. Initially, laboratory dissection measurements seed the placement of histology slices in corresponding CT volumes, followed by in-plane point-based registration of bone in histology images to the bone in CT. Validation using six bisected canine humerus ex-vivo CT datasets indicated a plane misalignment of 0.19 ± 1.8 mm. User input sensitivity was assessed at 0.08 ± 0.2 mm for plane translation and 0-1.6° deviation. These results show a similar magnitude of error to related prostate histology co-location work. Although demonstrated with a femoral canine sarcoma tumour, this solution can be generalised to various orthopaedic geometries and sites. It supports high-fidelity histology image co-location to improve understanding of tissue characterisation accuracy in clinical radiology. This solution requires only minimal adjustment to routine workflows. By integrating histology insights earlier in the presentation-diagnosis-planning-surgery-recovery loop, this solution guides data co-location to support the continued evaluation of safe pre-surgical margins.

Depression diagnosis: EEG-based cognitive biomarkers and machine learning

Depression is a complex mental illness that has significant effects on people as well as society. The traditional techniques for the diagnosis of depression, along with the potential of nascent biomarkers especially EEG-based biomarkers, are studied. This review explores the significance of cognitive biomarkers, offering a comprehensive understanding of their role in the overall assessment of depression. It also investigates the effects of depression on various brain regions, outlines promising areas for future research, and emphasizes the importance of understanding the neurophysiological roots of depression. Furthermore, it elucidates how machine learning and deep learning models are integrated into EEG-based depression diagnosis, emphasizing their importance in optimizing personalized therapeutic protocols and improving diagnostic accuracy with EEG data analysis.

A software tool for fabricating phantoms mimicking human tissues with designated dielectric properties and frequency

OBJECTIVES

Dielectric materials play a crucial role in assessing and refining the measurement performance of dielectric properties for specific tasks. The availability of viable and standardized dielectric materials could greatly enhance medical applications related to dielectric properties. However, obtaining reliable phantoms with designated dielectric properties across a specified frequency range remains challenging. In this study, we propose software to easily determine the components of dielectric materials in the frequency range of 16 MHz to 3 GHz.

METHODS

A total of 184 phantoms were fabricated and measured using open-ended coaxial probe method. The relationship among dielectric properties, frequency, and the components of dielectric materials was fitted through feedforward neural networks. Software was developed to quickly calculate the composition of dielectric materials.

RESULTS

We performed validation experiments including blood, muscle, skin, and lung tissue phantoms at 128 MHz, 298 MHz, 915 MHz, and 2.45 GHz. Compared with literature values, the relative errors of dielectric properties are less than 15 %.

CONCLUSIONS

This study establishes a reliable method for fabricating dielectric materials with designated dielectric properties and frequency through the development of the software. This research holds significant importance in enhancing medical research and applications that rely on tissue simulation using dielectric phantoms.

DeepCOVIDNet-CXR: deep learning strategies for identifying COVID-19 on enhanced chest X-rays

OBJECTIVES

COVID-19 is one of the recent major epidemics, which accelerates its mortality and prevalence worldwide. Most literature on chest X-ray-based COVID-19 analysis has focused on multi-case classification (COVID-19, pneumonia, and normal) by the advantages of Deep Learning. However, the limited number of chest X-rays with COVID-19 is a prominent deficiency for clinical relevance. This study aims at evaluating COVID-19 identification performances using adaptive histogram equalization (AHE) to feed the ConvNet architectures with reliable lung anatomy of airways.

METHODS

We experimented with balanced small- and large-scale COVID-19 databases using left lung, right lung, and complete chest X-rays with various AHE parameters. On multiple strategies, we applied transfer learning on four ConvNet architectures (MobileNet, DarkNet19, VGG16, and AlexNet).

RESULTS

Whereas DarkNet19 reached the highest multi-case identification performance with an accuracy rate of 98.26 % on the small-scale dataset, VGG16 achieved the best generalization performance with an accuracy rate of 95.04 % on the large-scale dataset.

CONCLUSIONS

Our study is one of the pioneering approaches that analyses 3615 COVID-19 cases and specifies the most responsible AHE parameters for ConvNet architectures in the multi-case classification.