Image Acquisition Research Articles

Three-Dimensional Morphological Characterisation of Human Cortical Organoids Using a Customised Image Analysis Workflow

Summary Statement: A tailored image analysis workflow was applied to quantify cortical organoid health, development, morphology and cellular composition over time. The assessment of cellular composition and viability of stem cell-derived organoid models is a complex but essential approach to understanding the mechanisms of human development and disease. Aim: Our study was motivated by the need for an image-analysis workflow, including high-cell content, high-throughput methods, to measure the architectural features of developing organoids. We assessed stem cell-derived cortical organoids at 4 and 6 months post-induction using immunohistochemistry-labelled sections as the analysis testbed. The workflow leveraged fluorescence imaging tailored to classify cells as viable and dying or non-viable and assign neuronal and astrocytic perinuclear markers to count cells. Results/Outcomes: Image acquisition was accelerated by capturing the organoid slice in 3D using widefield-fluorescence microscopy. This method used computational clearing to resolve nuclear and perinuclear markers and retain their spatial information within the organoid’s heterogeneous structure. The customised workflow analysed over 1.5 million cells using DAPI-stained nuclei, filtering and quantifying viable and non-viable cells and the necrotic-core regions. Temporal analyses of neuronal cell number derived from perinuclear labelling were consistent with organoid maturation from 4 to 6 months of in vitro differentiation. Overall: We have provided a comprehensive and enhanced image analysis workflow for organoid structural evaluation, creating the ability to gather cellular-level statistics in control and disease models.

Improving Industrial Quality Control: A Transfer Learning Approach to Surface Defect Detection

To automate the quality control of painted surfaces of heating devices, an automatic defect detection and classification system was developed by combining deflectometry and bright light-based illumination on the image acquisition, deep learning models for the classification of non-defective (OK) and defective (NOK) surfaces that fused dual-modal information at the decision level, and an online network for information dispatching and visualization. Three decision-making algorithms were tested for implementation: a new model built and trained from scratch and transfer learning of pre-trained networks (ResNet-50 and Inception V3). The results revealed that the two illumination modes employed widened the type of defects that could be identified with this system, while maintaining its lower computational complexity by performing multi-modal fusion at the decision level. Furthermore, the pre-trained networks achieved higher accuracies on defect classification compared to the self-built network, with ResNet-50 displaying higher accuracy. The inspection system consistently obtained fast and accurate surface classifications because it imposed OK classification on models trained with images from both illumination modes. The obtained surface information was then successfully sent to a server to be forwarded to a graphical user interface for visualization. The developed system showed considerable robustness, demonstrating its potential as an efficient tool for industrial quality control.

Simultaneous multislice diffusion imaging using navigator-free multishot spiral acquisitions.

This work aims to raise a novel design for navigator-free multiband (MB) multishot uniform-density spiral (UDS) acquisition and reconstruction, and to demonstrate its utility for high-efficiency, high-resolution diffusion imaging. Our design focuses on the acquisition and reconstruction of navigator-free MB multishot UDS diffusion imaging. For acquisition, radiofrequency-pulse encoding was used to achieve controlled aliasing in parallel imaging in MB imaging. For reconstruction, a new algorithm named slice-projection onto convex sets-enhanced inherent correction of phase errors (slice-POCS-ICE) was proposed to simultaneously estimate diffusion-weighted images and intershot phase variations for each slice. The efficacy of the proposed methods was evaluated in both numerical simulation and in vivo experiments. In both numerical simulation and in vivo experiments, slice-POCS-ICE estimated phase variations more precisely and provided results with better image quality than other methods. The intershot phase variations and MB slice aliasing artifacts were simultaneously resolved using the proposed slice-POCS-ICE algorithm. The proposed navigator-free MB multishot UDS acquisition and reconstruction method is an effective solution for high-efficiency, high-resolution diffusion imaging.

Remote Mapping of Bedrock for Future Cosmogenic Nuclide Exposure Dating Studies in Unvisited Areas of Antarctica

Cosmogenic nuclide exposure dating is an important technique for reconstructing glacial histories. Many of the most commonly applied cosmogenic nuclides are extracted from the mineral quartz, meaning sampling of felsic (silica-rich) rock is often preferred to sampling of mafic (silica-poor) rock for exposure dating studies. Fieldwork in remote regions such as Antarctica is subject to time constraints and considerable logistical challenges, making efficient sample recovery critical to successful research efforts. Remote sensing offers an effective way to map the geology of large areas prior to fieldwork and expedite the sampling process. In this study, we assess the viability of multispectral remote sensing to distinguish felsic from mafic rock outcrops at visible-near infrared (VNIR) and shortwave infrared (SWIR) wavelengths using both the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and very high-resolution Worldview-3 (WV-3) imagery. We applied a combination of spectral mapping and ground truth from spectral measurements of 17 rock samples from Mount Murphy in the Amundsen Sea sector of West Antarctica. Using this approach, we identified four dominant rock types which we used as a basis for felsic–mafic differentiation: felsic granites and gneisses, and mafic basalts and fragmental hydrovolcanic rocks. Supervised classification results indicate WV-3 performs well at differentiating felsic and mafic rock types and that ASTER, while coarser, could also achieve satisfactory results and be used in concert with more targeted WV-3 image acquisitions. Finally, we present a revised felsic–mafic geological map for Mt Murphy. Overall, our results highlight the potential of spectral mapping for preliminary reconnaissance when planning future cosmogenic nuclide sampling campaigns in remote, unvisited areas of the polar regions.

Assessment of the Quality of Root Canal Fillings-An Ex-Vivo Comparison of CBCT Scans, Conventional Intraoral Sensors, and a Novel Photon-Counting Sensor.

To compare a novel photon-counting sensor, two CBCT protocols and two CMOS sensors on the detection of gaps between a gutta-percha cone and root canal walls. Twenty-five mandibular incisors were prepared to 45/.04 (size/taper) at working length. Teeth were placed in a partially dentate mandible and single gutta-percha cones of seven sizes were placed at length, one at a time, for image acquisition with a photon-counting sensor, two CBCT protocols (90µm3, 120µm3) and two CMOS sensors. Three calibrated observers assessed images for gap presence. Sensitivity, specificity, accuracy, AUC, and agreement with gold standard were determined using ANOVA and Tukey test (p ≤ 0.05). Photon-counting sensor showed superior sensitivity and accuracy (88.47%, 81.57%), significantly higher than the CBCT protocols (50.70-56.33%, 45.87-53.17%). Contrarily, the photon-counting sensor showed the lowest specificity (40.27%), significantly lower than the CBCT protocols (90.27%, 97.23%). CMOS sensors showed sensitivity, specificity, and accuracy between 72.23-74.53%, not differing from other modalities. All intraoral sensors showed AUC around 82.87-84.03%, significantly higher than CBCT protocol 120µm3 (74.07%). The file size was inversely related to gap size and percentual agreement with gold standard. CMOS sensors showed consistent results, while the photon-counting sensor had the highest sensitivity but lacked specificity. CBCT protocols excelled in specificity but had lower sensitivity. Novel photon-counting sensors and CBCT imaging provided no significant advantage over conventional sensors in assessing gaps as an indicator of quality of root canal filling. Furthermore, smaller gaps were more difficult to detect, regardless of the imaging technique used.

Artificial Intelligence–Guided Lung Ultrasound by Nonexperts

Lung ultrasound (LUS) aids in the diagnosis of patients with dyspnea, including those with cardiogenic pulmonary edema, but requires technical proficiency for image acquisition. Previous research has demonstrated the effectiveness of artificial intelligence (AI) in guiding novice users to acquire high-quality cardiac ultrasound images, suggesting its potential for broader use in LUS. To evaluate the ability of AI to guide acquisition of diagnostic-quality LUS images by trained health care professionals (THCPs). In this multicenter diagnostic validation study conducted between July 2023 and December 2023, participants aged 21 years or older with shortness of breath recruited from 4 clinical sites underwent 2 ultrasound examinations: 1 examination by a THCP operator using Lung Guidance AI and the other by a trained LUS expert without AI. The THCPs (including medical assistants, respiratory therapists, and nurses) underwent standardized AI training for LUS acquisition before participation. Lung Guidance AI software uses deep learning algorithms guiding LUS image acquisition and B-line annotation. Using an 8-zone LUS protocol, the AI software automatically captures images of diagnostic quality. The primary end point was the proportion of THCP-acquired examinations of diagnostic quality according to a panel of 5 masked expert LUS readers, who provided remote review and ground truth validation. The intention-to-treat analysis included 176 participants (81 female participants [46.0%]; mean [SD] age, 63 [14] years; mean [SD] body mass index, 31 [8]). Overall, 98.3% (95% CI, 95.1%-99.4%) of THCP-acquired studies were of diagnostic quality, with no statistically significant difference in quality compared to LUS expert-acquired studies (difference, 1.7%; 95% CI, -1.6% to 5.0%). In this multicenter validation study, THCPs with AI assistance achieved LUS images meeting diagnostic standards compared with LUS experts without AI. This technology could extend access to LUS to underserved areas lacking expert personnel. ClinicalTrials.gov Identifier: NCT05992324.

Digitalization of the Workflow for Drone-Assisted Inspection and Automated Assessment of Industrial Buildings for Effective Maintenance Management

Industrial buildings are a key element in the industrial fabric, and their maintenance is essential to ensure their proper functioning and avoid disruptions and costly economic losses. Continuous maintenance based on an accurate diagnosis makes it possible to meet the challenges of aging infrastructures, which demands a reliable data-based assessment for maintenance management implementing corrective and preventive actions, according to the damage criticality. This paper researches an innovative digitalized process for the inspection and diagnosis of industrial buildings, which leads to categorizing and prioritizing maintenance actions in an objective and cost-effective way from the inspection data. The process integrates some technical developments carried out in this work, aimed to automate the workflow: the drone-based inspection, the building condition assessment from the definition of a standardized construction pathology library, and a visual analysis of pathology evolution based on photogrammetry. The use of drones for digitalized inspection involves some challenges related to the positioning of the drone for damage localization, which has been herein overcome by developing a geo-annotation system for image acquisition. This system has also enabled the capture of geo-located images intended to generate 3D photogrammetric models for quantifying the pathological process evolution. Moreover, the assessment procedure outlined through multi-criteria decision-making methodology MIVES establishes a single criterion to automatically weight the relative importance of the damage defined in the library. As a result, this procedure yields the so-called Intervention Urgency Index (IUI), which allows prioritizing the maintenance actions associated with the damage while also considering economic criteria. In such a way, the overall process aims to increase reliability and consistency in the results of inspection and diagnosis needed for the effective maintenance management of industrial buildings.

Research on Innovative Apple Grading Technology Driven by Intelligent Vision and Machine Learning

In the domain of food science, apple grading holds significant research value and application potential. Currently, apple grading predominantly relies on manual methods, which present challenges such as low production efficiency and high subjectivity. This study marks the first integration of advanced computer vision, image processing, and machine learning technologies to design an innovative automated apple grading system. The system aims to reduce human interference and enhance grading efficiency and accuracy. A lightweight detection algorithm, FDNet-p, was developed to capture stem features, and a strategy for auxiliary positioning was designed for image acquisition. An improved DPC-AWKNN segmentation algorithm is proposed for segmenting the apple body. Image processing techniques are employed to extract apple features, such as color, shape, and diameter, culminating in the development of an intelligent apple grading model using the GBDT algorithm. Experimental results demonstrate that, in stem detection tasks, the lightweight FDNet-p model exhibits superior performance compared to various detection models, achieving an mAP@0.5 of 96.6%, with a GFLOPs of 3.4 and a model size of just 2.5 MB. In apple grading experiments, the GBDT grading model achieved the best comprehensive performance among classification models, with weighted Jacard Score, Precision, Recall, and F1 Score values of 0.9506, 0.9196, 0.9683, and 0.9513, respectively. The proposed stem detection and apple body classification models provide innovative solutions for detection and classification tasks in automated fruit grading, offering a comprehensive and replicable research framework for standardizing image processing and feature extraction for apples and similar spherical fruit bodies.

Predictive value of CT-based and AI-reconstructed 3D-TAPSE in patients undergoing transcatheter tricuspid valve repair

BackgroundThe tricuspid annular plane systolic excursion (TAPSE) assessed by echocardiography has failed in predicting outcomes in patients with severe tricuspid regurgitation (TR) undergoing transcatheter tricuspid valve intervention (TTVI). Considering the complex shape of the tricuspid annulus and right ventricle, as well as the difficult echocardiographic image acquisition of the right heart, cardiac computed tomography (CT) might be superior for the analysis of the annular excursion. Thus, this study aimed to analyze whether CT-captured TAPSE provides additional value in predicting outcomes after TTVI.Methods and resultsFor TTVI procedure planning, 75 patients (mean age, 77 ± 8 years; 61% female) with severe TR underwent full cardiac cycle CT. Septal, lateral, anterior, and posterior TAPSE, as well as TAPSE- volume, were analyzed. Indexed anterior and posterior (iTAPSE) and TAPSE volume were reduced in patients with right ventricular ejection fraction <45%. At 1 year after TTVI (mean follow-up, 193 ± 146days), the combined endpoint of death and rehospitalization occurred in significantly fewer patients with posterior iTAPSE >4.5 mm/m2 (17.2% vs. 63.6%; HR 0.225, CI 0.087–0.581; P < 0.001) and in patients with iTAPSE volume >9 ml/m2 (16.4% vs. 57.1%; HR: 0.269 CI 0.105–0.686; P = 0.003). Echocardiographic TAPSE correlated best with lateral CT-based TAPSE, although both failed in predicting outcomes after TTVI. In multivariate Cox regression, posterior iTAPSE was found to be a significant predictor of outcome 1 year after TTVI.ConclusionsPosterior iTAPSE is an independent predictor of cardiovascular outcomes among patients undergoing TTVI. Furthermore, CT-measured TAPSE has incremental value and refines risk stratification for clinical outcomes in patients undergoing TTVI.

Barley leaf cell-wall responses to the penetration by adapted and nonadapted powdery mildew fungi revealed with the thermal-source based Fourier transform infrared microspectroscopy and focal plane array

Infrared microspectroscopy allows for detailed characterization of functional groups at subcellular levels in plants. In this study, barley leaves were inoculated with Blumeria graminis f. sp. hordei (Bgh, adapted) and B. graminis f. sp. tritici (Bgt, nonadapted). Globar-based transmission Fourier transform infrared (gFTIR) microspectroscopy equipped with an upgraded high-resolution focal plane array (FPA) detector which enabled the rapid acquisition of high magnification imaging with a 3.3-μm × 3.3-μm capacity, was employed to quantitatively analyse the cell wall (CW) modification over individual leaf cells in response to attempted penetration at 24 hours post-inoculation. Univariate and multivariate analyses of gFTIR spectra revealed distinct changes in several CW functional groups upon attempted penetration by Bgt, with significant increases in the integrated areas of cellulose, hemicellulose, methyl-esterified pectins and lipid regions compared to cells without penetration. Notably, neither guaiacyl (G) nor syringyl (S) lignin increased significantly in CWs near Bgt penetration sites relative to nonpenetrating controls, though the G/S ratio was higher than in Bgh-penetrated sites. These findings suggest early and rapid CW modifications occur during the nonhost resistance to Bgt. This study also confirms previous results obtained using synchrotron-based FTIR (sFTIR), demonstrating that both gFTIR and sFTIR are valuable for studying plant CW apposition, with gFTIR being more accessible and rapid for imaging over large areas with compatible quality and resolution to sFTIR.

Rapid Acquisition of High-Pixel Fluorescence Lifetime Images of Living Cells via Image Reconstruction Based on Edge-Preserving Interpolation

Fluorescence lifetime imaging (FLIM) has established itself as a pivotal tool for investigating biological processes within living cells. However, the extensive imaging duration necessary to accumulate sufficient photons for accurate fluorescence lifetime calculations poses a significant obstacle to achieving high-resolution monitoring of cellular dynamics. In this study, we introduce an image reconstruction method based on the edge-preserving interpolation method (EPIM), which transforms rapidly acquired low-resolution FLIM data into high-pixel images, thereby eliminating the need for extended acquisition times. Specifically, we decouple the grayscale image and the fluorescence lifetime matrix and perform an individual interpolation on each. Following the interpolation of the intensity image, we apply wavelet transformation and adjust the wavelet coefficients according to the image gradients. After the inverse transformation, the original image is obtained and subjected to noise reduction to complete the image reconstruction process. Subsequently, each pixel is pseudo-color-coded based on its intensity and lifetime, preserving both structural and temporal information. We evaluated the performance of the bicubic interpolation method and our image reconstruction approach on fluorescence microspheres and fixed-cell samples, demonstrating their effectiveness in enhancing the quality of lifetime images. By applying these techniques to live-cell imaging, we can successfully obtain high-pixel FLIM images at shortened intervals, facilitating the capture of rapid cellular events.

Fast Hadamard-Encoded 7T Spectroscopic Imaging of Human Brain

Background/Objectives: The increased SNR available at 7T combined with fast readout trajectories enables accelerated spectroscopic imaging acquisitions for clinical applications. In this report, we evaluate the performance of a Hadamard slice encoding strategy with a 2D rosette trajectory for multi-slice fast spectroscopic imaging at 7T. Methods: Moderate-TE (~40 ms) spin echo and J-refocused polarization transfer sequences were acquired with simultaneous Hadamard multi-slice excitations and rosette in-plane encoding. The moderate spin echo sequence, which targets singlet compounds (i.e., N-acetyl aspartate, creatine, and choline), uses cascaded multi-slice RF excitation pulses to minimize the chemical shift dispersion error. The J-refocused sequence targets coupled spin systems (i.e., glutamate and myo-inositol) using simultaneous multi-slice excitation to maintain the same TE across all slices. A modified Hadamard slice encoding strategy was used to decrease the peak RF pulse amplitude of the simultaneous multi-slice excitation pulse for the J-refocused acquisition. Results: The accuracy of multi-slice and single-slice rosette spectroscopic imaging (RSI) is comparable to conventional Cartesian-encoded spectroscopic imaging (CSI). Spectral analyses for the J-refocused studies of glutamate and myo-inositol show that the Cramer Rao lower bounds are not significantly different between the fast RSI and conventional CSI studies. Linear regressions of creatine/N-acetyl aspartate and glutamate/N-acetyl aspartate with tissue gray matter content are consistent with literature values. Conclusions: With minimal gradient demands and fast acquisition times, the 2.2 min to 9 min for single- to four-slice RSI acquisitions are well tolerated by healthy subjects and tumor patients, and show results that are consistent with clinical outcomes.

Novel RFID Multi-Label Network Location Measurement by Dual CCD and Non-Local Means–Harris Algorithm

To improve the performance of Radio Frequency Identification (RFID) multi-label systems, the multi-label network structure needs to be quickly located and optimized. A multi-label location measurement method based on the NLM–Harris algorithm is proposed in this paper. Firstly, multi-label geometric distribution images are obtained through a label image acquisition system of a multi-label semi-physical simulation platform with two vertical Charge-Coupled Device (CCD) cameras, and Gaussian noise is added to the image to simulate thermoelectric interference. Then, a fast NLM algorithm that optimizes the kernel coefficient acquisition speed is used for image denoising. Finally, the Harris corner algorithm is used to obtain the corner points of the images. After screening the diagonal points, the pixel coordinates of the preset origin and the four corners of the labels are obtained. Furthermore, the actual coordinates of the labels are obtained according to the pixel relationship. The results show that the average absolute errors of x, y, and z coordinates are 0.773 mm, 0.782 mm, and 0.807 mm, respectively. In addition, the relative errors are 1.659%, 2.260%, and 0.258%, which shows the high location accuracy of the multi-label network. It is of great significance to measure and optimize the performance of multi-label systems.

The textures of sarcoidosis: quantifying lung disease through variograms

Objective. Sarcoidosis is a granulomatous disease affecting the lungs in over 90% of patients. Qualitative assessment of chest CT by radiologists is standard clinical practice and reliable quantification of disease from CT would support ongoing efforts to identify sarcoidosis phenotypes. Standard imaging feature engineering techniques such as radiomics suffer from extreme sensitivity to image acquisition and processing, potentially impeding generalizability of research to clinical populations. In this work, we instead investigate approaches to engineering variogram-based features with the intent to identify a robust, generalizable pipeline for image quantification in the study of sarcoidosis.Approach. For a cohort of more than 300 individuals with sarcoidosis, we investigated 24 feature engineering pipelines differing by decisions for image registration to a template lung, empirical and model variogram estimation methods, and feature harmonization for CT scanner model, and subsequently 48 sets of phenotypes produced through unsupervised clustering. We then assessed sensitivity of engineered features, phenotypes produced through unsupervised clustering, and sarcoidosis disease signal strength to pipeline.Main results. We found that variogram features had low to mild association with scanner model and associations were reduced by image registration. For each feature type, features were also typically robust to all pipeline decisions except image registration. Strength of disease signal as measured by association with pulmonary function testing and some radiologist visual assessments was strong (optimistic AUC ≈ 0.9,p≪0.0001in models for architectural distortion, conglomerate mass, fibrotic abnormality, and traction bronchiectasis) and fairly consistent across engineering approaches regardless of registration and harmonization for CT scanner.Significance. Variogram-based features appear to be a suitable approach to image quantification in support of generalizable research in pulmonary sarcoidosis.

Optimizing hip MRI: enhancing image quality and elevating inter-observer consistency using deep learning-powered reconstruction

BackgroundConventional hip joint MRI scans necessitate lengthy scan durations, posing challenges for patient comfort and clinical efficiency. Previously, accelerated imaging techniques were constrained by a trade-off between noise and resolution. Leveraging deep learning-based reconstruction (DLR) holds the potential to mitigate scan time without compromising image quality.MethodsWe enrolled a cohort of sixty patients who underwent DL-MRI, conventional MRI, and No-DL MRI examinations to evaluate image quality. Key metrics considered in the assessment included scan duration, overall image quality, quantitative assessments of Relative Signal-to-Noise Ratio (rSNR), Relative Contrast-to-Noise Ratio (rCNR), and diagnostic efficacy. Two experienced radiologists independently assessed image quality using a 5-point scale (5 indicating the highest quality). To gauge interobserver agreement for the assessed pathologies across image sets, we employed weighted kappa statistics. Additionally, the Wilcoxon signed rank test was employed to compare image quality and quantitative rSNR and rCNR measurements.ResultsScan time was significantly reduced with DL-MRI and represented an approximate 66.5% reduction. DL-MRI consistently exhibited superior image quality in both coronal T2WI and axial T2WI when compared to both conventional MRI (p < 0.01) and No-DL-MRI (p < 0.01). Interobserver agreement was robust, with kappa values exceeding 0.735. For rSNR data, coronal fat-saturated(FS) T2WI and axial FS T2WI in DL-MRI consistently outperformed No-DL-MRI, with statistical significance (p < 0.01) observed in all cases. Similarly, rCNR data revealed significant improvements (p < 0.01) in coronal FS T2WI of DL-MRI when compared to No-DL-MRI. Importantly, our findings indicated that DL-MRI demonstrated diagnostic performance comparable to conventional MRI.ConclusionIntegrating deep learning-based reconstruction methods into standard clinical workflows has the potential to the promise of accelerating image acquisition, enhancing image clarity, and increasing patient throughput, thereby optimizing diagnostic efficiency.Trial registrationRetrospectively registered.

Real-Time Deployment of Ultrasound Image Interpretation AI Models for Emergency Medicine Triage Using a Swine Model

Ultrasound imaging is commonly used for medical triage in both civilian and military emergency medicine sectors. One specific application is the eFAST, or the extended focused assessment with sonography in trauma exam, where pneumothorax, hemothorax, or abdominal hemorrhage injuries are identified. However, the diagnostic accuracy of an eFAST exam depends on obtaining proper scans and making quick interpretation decisions to evacuate casualties or administer necessary interventions. To improve ultrasound interpretation, we developed AI models to identify key anatomical structures at eFAST scan sites, simplifying image acquisition by assisting with proper probe placement. These models plus image interpretation diagnostic models were paired with two real-time eFAST implementations. The first implementation was a manual AI-driven ultrasound eFAST tool that used guidance models to select correct frames prior to making any diagnostic predictions. The second implementation was a robotic imaging platform capable of providing semi-autonomous image acquisition combined with diagnostic image interpretation. We highlight the use of both real-time approaches in a swine injury model and compare their performance of this emergency medicine application. In conclusion, AI can be deployed in real time to provide rapid triage decisions, lowering the skill threshold for ultrasound imaging at or near the point of injury.

Staging of prostate Cancer with ultra-fast PSMA-PET scans enhanced by AI.

PSMA-PET is a reference standard examination for patients with prostate cancer, but even using recently introduced digital PET detectors image acquisition with standard field-of-view scanners is still in the range of 20min. This may cause limited access to examination slots because of the growing demand for PSMA-PET. Ultra-fast PSMA-PET may enhance throughput but comes at the cost of poor image quality. The aim of this manuscript is to evaluate the accuracy of AI-enhanced ultra-fast PSMA-PET for staging of patients with prostate cancer. A total number of 357 whole-body [68Ga]Ga-PSMA-11 PET datasets were included. Patients underwent two digital PET scans, one at standard and one at ultra-fast speed (table speed: 0.6-1.2mm/s vs. 50mm/s). A modified pix2pixHD generative adversarial network to enhance the ultra-fast images was trained with 286 datasets and evaluated with the remaining 71 datasets. The staging accuracy of ultra-fast PSMA-PET and AI-enhanced ultra-fast PET was compared with the reference standard PET separately for miTNM regions proposed by PROMISE V2.0. The AI-network significantly improved the visual image quality and detection rate in most miTNM regions compared with the non-enhanced image data (T: 69.6% vs. 43.5%, p < 0.05; N: 46.3% vs. 27.8%, p < 0.01; M1a 64.4% vs. 47.5%, p < 0.01; M1b: 85.7% vs. 72.1%, p < 0.01). However, improvement was not significant for the M1c category (42.9 vs. 28.6%, p > 0.05). Missed lesions had a smaller SUVmax and lesion size compared with detected lesions (exemplary for N: 9.5 vs. 26.5 SUVmax; 4 vs. 10mm). SUVmax values of lesions were significantly different in all miTNM regions between the ultra-fast and reference standard PET, but only in the T-region between the AI-enhanced and reference standard PET. The AI-based image enhancement improved image quality and region detection rates by a mean of 17.9%. As the sensitivity of synthetic PET for small and low-uptake lesions was limited, a potential clinical use case could be disease monitoring in patients with high tumor volume and PSMA uptake undergoing PSMA radioligand therapy. The improvement in detection rate of distant metastases was not significant. This indicates that more training data is needed to ensure robust results also for lesions that have lower appearance frequency. Future studies on accelerated PSMA-PET seem warranted.

Protocol for the three-dimensional analysis of rodent skeletal muscle.

Confocal imaging is a powerful tool capable of analyzing cellular spatial data within a given tissue. Here, we present a protocol for preparing optically cleared extensor digitorum longus (EDL) skeletal muscle samples suitable for confocal imaging/computational analysis. We describe steps for sample preparation (including perfusion fixation and tissue clearing of muscle samples), image acquisition, and computational analysis, with sample segmentation/3D rendering outlined. This protocol can be applied to characterize various cell types, including muscle satellite cells (muscle stem cells) and capillary endothelial cells within rodent skeletal muscle. For complete details on the use and execution of this protocol, please refer to Verma etal.,1 Verma etal.,2 Karthikeyan etal.,3 and Karthikeyan et al.4.

Enhance registration precision of transmission breast images utilizing improved Levenberg-Marquardt optimization algorithm with normalized cross-correlation.

Transmission imaging may become a possible advance for breast cancer screening with non-invasive, cost-effective, and radiation-free approaches for early detection. Frame accumulation can successfully eliminate the issue of low SNR, low grayscale and poor quality in transmission image. However, frame accumulation accuracy can be diminished because of inherent human body instability during image acquisition and the light absorption characteristics of breast tissue, resulting in distorted and misplaced image sequences. Therefore, improved Levenberg-Marquardt optimization algorithm with normalized cross-correlation is used as an innovative approach to rectify image sequences before frame accumulation processing. Two separate sets of data, showing breast images with and without markers, were collected using a halogen bulb and a mobile phone camera to validate the suggested method. The approach includes coarse registration utilizing normalized cross-correlation for initial value estimation, followed by fine registration using Levenberg-Marquardt algorithm. The results demonstrate a notable improvement in both accuracy of registration and frame accumulation quality. Specifically, the registration speed showed a remarkable increase, being 8.7 times faster, especially prominent in images that included markers. These images displayed normalized cross-correlation values reaching up to 0.99. The research emphasizes the future potential of the suggested method in overcoming the image quality challenges associated with breast transmission imaging, providing a significant milestone toward more accurate and efficient early breast cancer screening methods. Moreover, transmission imaging systems for the breast have been developed to verify the safety and effectiveness of the implemented technology.

Artificial Intelligence in Radiology: A Leadership Survey.

Surveys to assess views about artificial intelligence (AI) of various diagnostic radiology constituencies have revealed interesting combinations of enthusiasm, caution, and implementation priorities. We surveyed academic radiology leaders about their views on AI and how they intend to approach AI implementation in their departments. We conducted a web survey of Society of Chairs of Academic Radiology Departments (SCARD) members between October 5 and October 31, 2023 to solicit optimism or pessimism about AI, target use cases, planned implementation, and perceptions of their workforce. P-values are provided only for descriptive purposes and have not been adjusted for multiple testing in this exploratory research. The survey was sent to the 112 SCARD members and 43 responded (38%). Chairs were optimistic, with no statistical difference between views of AI in general versus generative AI. Chairs plan to implement AI to improve quality and efficiency (43/43, 100%), burnout (41/43, 95%), healthcare costs (22/43, 51%), and equity (27/43, 63%) and most likely will target the post-processing (26/43, 60%), interpretation workflow (26/43, 60%), and image acquisition (18/43, 42%) steps in the imaging value chain. Chairs perceived that radiologists (36/43, 84%) and technologists (38/43, 88%) were not particularly worried about being displaced but saw trainees as slightly less confident (31/43, 72%). Free text responses revealed concerns about the cost of AI and emphasized trade-offs that needed to be balanced. Radiology Chairs are optimistic about AI and poised to tackle departmental challenges. Concerns about generative AI and workforce replacement are minimal.