Sensitivity For Detection Of Lesions Research Articles

Deep Learning-Accelerated Liver Diffusion-Weighted Imaging: Intraindividual Comparison and Additional Phantom Study of Free-Breathing and Respiratory-Triggering Acquisitions.

Deep learning-reconstructed diffusion-weighted imaging (DL-DWI) is an emerging promising time-efficient method for liver evaluation, but analyses regarding different motion compensation strategies are lacking. This study evaluated the qualitative and quantitative features, sensitivity for focal lesion detection, and scan time of free-breathing (FB) DL-DWI and respiratory-triggered (RT) DL-DWI compared with RT conventional DWI (C-DWI) in the liver and a phantom. Eighty-six patients indicated for liver MRI underwent RT C-DWI, FB DL-DWI, and RT DL-DWI with matching imaging parameters other than the parallel imaging factor and number of averages. Two abdominal radiologists independently assessed qualitative features (structural sharpness, image noise, artifacts, and overall image quality) using a 5-point scale. The signal-to-noise ratio (SNR) along with the apparent diffusion coefficient (ADC) value and its standard deviation (SD) were measured in the liver parenchyma and a dedicated diffusion phantom. For focal lesions, per-lesion sensitivity, conspicuity score, SNR, and ADC value were evaluated. Wilcoxon signed rank test and repeated-measures analysis of variance with post hoc test revealed the difference in DWI sequences. Compared with RT C-DWI, the scan times for FB DL-DWI and RT DL-DWI were reduced by 61.5% and 23.9%, respectively, with statistically significant differences between all 3 pairs (all P's < 0.001). Respiratory-triggered DL-DWI showed a significantly sharper liver margin, less image noise, and more minor cardiac motion artifact compared with RT C-DWI (all P's < 0.001), whereas FB DL-DWI showed more blurred liver margins and poorer intrahepatic vessels demarcation than RT C-DWI. Both FB- and RT DL-DWI showed significantly higher SNRs than RT C-DWI in all liver segments (all P's < 0.001). There was no significant difference in overall ADC values across DWI sequences in the patient or phantom, with the highest value recorded in the left liver dome by RT C-DWI. The overall SD was significantly lower with FB DL-DWI and RT DL-DWI than RT C-DWI (all P's ≤ 0.003). Respiratory-triggered DL-DWI showed a similar per-lesion sensitivity (0.96; 95% confidence interval, 0.90-0.99) and conspicuity score to those of RT C-DWI and significantly higher SNR and contrast-to-noise ratio values (P ≤ 0.006). The per-lesion sensitivity of FB DL-DWI (0.91; 95% confidence interval, 0.85-0.95) was significantly lower than that of RT C-DWI (P = 0.001), with a significantly lower conspicuity score. Compared with RT C-DWI, RT DL-DWI demonstrated superior SNR, comparable sensitivity for focal hepatic lesions, and reduced acquisition time, making it a suitable alternative to RT C-DWI. Despite FB DL-DWI's weakness in motion-related challenges, further refinement could potentiate FB DL-DWI in the context of abbreviated screening protocols, where time efficiency is a high priority.

An Australian experience using Tc-PSMA SPECT/CT in the primary diagnosis of prostate cancer and for staging at biochemical recurrence after local therapy.

Technetium 99 prostate-specific membrane antigen(Tc-PSMA) single-photon emission computed tomography/computed tomography (SPECT/CT) has the potential to provide greater accessibility globally thangallium 68(Ga)-PSMA positron emission tomography (PET)/CT but has not been studied as extensively in primary diagnosis, staging, or relapse of prostate cancer (PC). We instituted a novel SPECT/CT reconstruction algorithm using Tc-PSMA and established a database to prospectively accumulate data on all patients referred with PC. This study extracts data on all patients referred over a 3.5-year period with the primary aim of comparing the diagnostic accuracy of Tc-PSMA and multiparametric magnetic resonance imaging (mpMRI) in the primary diagnosis of PC. The secondary aim was to assess the sensitivity of Tc-PSMA in detecting disease with relapse after either radical prostatectomy or primary radiotherapy. A total of 425 men referred for primary staging (PS) of PC and 172 men referred with biochemical relapse (BCR) were evaluated. We evaluated diagnostic accuracy and correlations between Tc-PSMA SPECT/CT, magnetic resonance imaging(MRI), prostate biopsy, prostate-specific antigen (PSA), and age in the PS group and positivity rates at different PSA levels in the BCR group. Taking the biopsy's grade according to the International Society of Urological Pathologyprotocol as a reference, the sensitivity (true positive rate), specificity (true negative rate), accuracy (positive and negative predictive value), and precision (positive predictive value) for Tc-PSMA in the PS group were 99.7%, 83.3%, 99.4%, and 99.7%, respectively. Comparison rates for MRI in this group were 96.4%, 71.4%, 95.7%, and 99.1%. We found moderate correlations between Tc-PSMA uptake in the prostate and biopsy grade, the presence of metastases, and PSA. In BCR, the Tc-PSMA positive rates were 38.9%, 53.2%, 62.5%, and 84.6% at PSA levels of <0.2, 0.2 to <0.5, 0.5 to <1.0, and > 1.0 ng/mL respectively. We have shown that Tc-PSMA SPECT/CT using an enhanced reconstruction algorithm has a diagnostic performance similar to Ga-PSMA PET/CT and mpMRI in an everyday clinical setting. It may have some advantages in cost, sensitivity for primary lesion detection, and the ability for intraoperative localization of lymph nodes.

An Assessment of the Pathological Classification and Postoperative Outcome of Focal Cortical Dysplasia by Simultaneous Hybrid PET/MRI

The purpose of this research was to investigate whether MRI and Simultaneous Hybrid PET/MRI images were consistent in the histological classification of patients with focal cortical dysplasia. Additionally, this research aimed to evaluate the postoperative outcomes with the MRI and Simultaneous Hybrid PET/MRI images of focal cortical dysplasia. A total of 69 cases in this research were evaluated preoperatively for drug-resistant seizures, and then surgical resection procedures of the epileptogenic foci were performed. The postoperative result was histopathologically confirmed as focal cortical dysplasia, and patients then underwent PET and MRI imaging within one month of the seizure. In this study, head MRI was performed using a 3.0 T magnetic resonance scanner (Philips) to obtain 3D T1WI images. The Siemens Biograph 16 scanner was used for a routine scanning of the head to obtain PET images. BrainLAB's iPlan software was used to fuse 3D T1 images with PET images to obtain PET/MRI images. Focal cortical dysplasia was divided into three types according to ILAE: three patients were classified as type I, twenty-five patients as type II, and forty-one patients as type III. Patients age of onset under 18 and age of operation over 18 had a longer duration (p = 0.036, p = 0.021). MRI had a high lesion detection sensitivity of type III focal cortical dysplasia (p = 0.003). Simultaneous Hybrid PET/MRI showed high sensitivity in detecting type II and III focal cortical dysplasia lesions (p = 0.037). The lesions in Simultaneous Hybrid PET/MRI-positive focal cortical dysplasia patients were mostly located in the temporal and multilobar (p = 0.005, 0.040). Simultaneous Hybrid PET/MRI has a high accuracy in detecting the classification of focal cortical dysplasia. The results of this study indicate that patients with focal cortical dysplasia with positive Simultaneous Hybrid PET/MRI have better postoperative prognoses.

Multi-arm U-Net with dense input and skip connectivity for T2 lesion segmentation in clinical trials of multiple sclerosis

T2 lesion quantification plays a crucial role in monitoring disease progression and evaluating treatment response in multiple sclerosis (MS). We developed a 3D, multi-arm U-Net for T2 lesion segmentation, which was trained on a large, multicenter clinical trial dataset of relapsing MS. We investigated its generalization to other relapsing and primary progressive MS clinical trial datasets, and to an external dataset from the MICCAI 2016 MS lesion segmentation challenge. Additionally, we assessed the model’s ability to reproduce the separation of T2 lesion volumes between treatment and control arms; and the association of baseline T2 lesion volumes with clinical disability scores compared with manual lesion annotations. The trained model achieved a mean dice coefficient of ≥ 0.66 and a lesion detection sensitivity of ≥ 0.72 across the internal test datasets. On the external test dataset, the model achieved a mean dice coefficient of 0.62, which is comparable to 0.59 from the best model in the challenge, and a lesion detection sensitivity of 0.68. Lesion detection performance was reduced for smaller lesions (≤ 30 μL, 3–10 voxels). The model successfully maintained the separation of the longitudinal changes in T2 lesion volumes between the treatment and control arms. Such tools could facilitate semi-automated MS lesion quantification; and reduce rater burden in clinical trials.

Autonomous Robotic System for Breast Biopsy With Deformation Compensation

Image-guided biopsy is the most common technique for breast cancer diagnosis. Although magnetic resonance imaging (MRI) has the highest sensitivity in breast lesion detection, ultrasound (US) biopsy guidance is generally preferred due to its non-invasiveness and real-time image feedback during the insertion. In this work, we propose an autonomous robotic system for US-guided biopsy of breast lesions identified on pre-operative MRI. After initial MRI to breast registration, the US probe attached to the robotic manipulator compresses the breast tissues until a pre-determined force level is reached. This technique, known as preloading, will allow to minimize lesion displacement during the needle insertion. Our workflow integrates a deformation compensation strategy based on patient-specific biomechanical model to update the US probe orientation keeping the target lesion on the US image plane during compression. By relying on a deformation model, the proposed system does not require lesion visibility on US. Experimental evaluation is performed to assess the performance of the system on a realistic breast phantom with 15 internal lesions, considering different preloading forces. The deformation compensation strategy allows to improve localization accuracy, and as a consequence final probe positioning, for all considered lesions. Median lesion localization error is 3.3 mm, which is lower than the median lesion radius, when applying a preloading of 2 N, which guarantees both minimal needle insertion error and tissue stress.

Design of a Low-Dose, Stationary, Tomographic Molecular Breast Imaging System Using 3-D Position Sensitive CZT Detectors

Molecular breast imaging (MBI) has been shown to have high sensitivity for lesion detection, particularly in patients with dense breasts where conventional mammography is limited. However, relatively high radiation dose and long imaging time are limiting factors. Most current MBI systems are based on planar imaging. Improved performance can be achieved using tomographic techniques, which normally involve detector motion. Our goal is to develop a low-dose stationary tomographic MBI system with similar or better performance in terms of lesion detection compared to planar MBI. The proposed system utilizes two opposing cadmium zinc telluride detectors with high intrinsic resolution and depth of interaction (DOI) capability, combined with densely packed multipinhole collimators. This leads to improved efficiency and adequate angular sampling, but also to significant multiplexing (MX), which can result in artefacts. We have developed de-MX algorithms that take advantage of the DOI information. We have performed both analytic and Monte Carlo simulations to demonstrate the feasibility, optimize the design and investigate the expected performance of the proposed system. Lesion detectability was preserved with reduction of acquisition time (or radiation dose) by a factor of 2 compared to planar images at the lowest reported dose. The first prototype is under evaluation at Kromek.

Comparison of Digital Radiography, Computed Tomography, and Magnetic Resonance Imaging Features in Canine Spontaneous Degenerative Stifle Joint Osteoarthritis

Canine stifle joint osteoarthritis (OA) is characterized by damage and degeneration of the articular cartilage and subchondral bone, bony hypertrophy at the margins, and synovial joint membrane changes. Non-invasive imaging modalities, such as digital radiography (DR), computed tomography (CT), and magnetic resonance imaging (MRI), can be used to describe these changes. However, the value of MRI in diagnosing spontaneous canine OA and the comparison of different imaging modalities have seldom been addressed. This study compared multiple noninvasive imaging modalities in canine spontaneous stifle OA cases. Four client-owned dogs with five spontaneously affected OA stifle joints were recruited and underwent DR, CT, and MRI. Information on osteophytes/enthesophytes, ligament/tendon lesions, synovial effusion and membrane thickening, subchondral bone lesions, and meniscal and cartilage lesions were scored and compared. The results showed that MRI provides the most comprehensive and superior lesion detection sensitivity for ligament, meniscus, cartilage, and synovial effusions. DR provides adequate bony structure information, while CT provides the most delicate images of bony structure lesions. These imaging findings may provide further understanding of the disease and help clinicians draft a more precise treatment plan.

Synthesis and Preclinical Evaluation of Three Novel 68Ga-Labeled Bispecific PSMA/FAP-Targeting Tracers for Prostate Cancer Imaging.

Tumor heterogeneity limits the efficacy and reliability of monospecific radiopharmaceuticals in prostate cancer diagnosis and therapy. To overcome this limitation and improve lesion detection sensitivity, we developed and evaluated three bispecific radiotracers that can target both prostate-specific membrane antigen (PSMA) and fibroblast activation protein (FAP), which are the two key proteins overexpressed in prostate cancer. Three FAP-targeting ligands with various linker lengths were synthesized through multistep organic synthesis, and then connected to the PSMA-targeting motif. IC50(PSMA) and IC50(FAP) values of Ga-complexed bispecific ligands, Ga-AV01017, Ga-AV01030, and Ga-AV01038 were 25.2-71.6 and 1.25-2.74 nM, respectively. The uptake values in PSMA-expressing LNCaP tumor xenografts were 4.38 ± 0.55, 5.17 ± 0.51, and 4.25 ± 0.86 %ID/g for [68Ga]Ga-AV01017, [68Ga]Ga-AV01030, and [68Ga]Ga-AV01038, respectively, which were lower than the monospecific PSMA-targeting tracer [68Ga]Ga-HTK03041 (23.1 ± 6.11 %ID/g). The uptake values in FAP-expressing HEK293T:hFAP tumor xenografts were 2.99 ± 0.37, 3.69 ± 0.81, 3.64 ± 0.83 %ID/g for [68Ga]Ga-AV01017, [68Ga]Ga-AV01030, and [68Ga]Ga-AV01038, respectively, which were also lower than the monospecific FAP-targeting tracer, [68Ga]Ga-FAPI-04 (12.5 ± 2.00 %ID/g). We observed that the bispecific tracers had prolonged blood retention, in which tracers with a longer linker tend to have a higher blood uptake and lower tumor uptake. Further investigations are needed to optimize the linker selection to generate promising bispecific PSMA/FAP-targeting tracers for prostate cancer imaging.

Automated quantification of PET/CT skeletal tumor burden in prostate cancer using artificial intelligence: The PET index

PurposeConsistent assessment of bone metastases is crucial for patient management and clinical trials in prostate cancer (PCa). We aimed to develop a fully automated convolutional neural network (CNN)-based model for calculating PET/CT skeletal tumor burden in patients with PCa.MethodsA total of 168 patients from three centers were divided into training, validation, and test groups. Manual annotations of skeletal lesions in [18F]fluoride PET/CT scans were used to train a CNN. The AI model was evaluated in 26 patients and compared to segmentations by physicians and to a SUV 15 threshold. PET index representing the percentage of skeletal volume taken up by lesions was estimated.ResultsThere was no case in which all readers agreed on prevalence of lesions that the AI model failed to detect. PET index by the AI model correlated moderately strong to physician PET index (mean r = 0.69). Threshold PET index correlated fairly with physician PET index (mean r = 0.49). The sensitivity for lesion detection was 65–76% for AI, 68–91% for physicians, and 44–51% for threshold depending on which physician was considered reference.ConclusionIt was possible to develop an AI-based model for automated assessment of PET/CT skeletal tumor burden. The model’s performance was superior to using a threshold and provides fully automated calculation of whole-body skeletal tumor burden. It could be further developed to apply to different radiotracers. Objective scan evaluation is a first step toward developing a PET/CT imaging biomarker for PCa skeletal metastases.

Head-to-head comparison of [18F]-FDG and [68Ga]-DOTA-FAPI-04 PET/CT for radiological evaluation of platinum-sensitive recurrent ovarian cancer.

Precise radiological evaluation is pivotal for the management of platinum-sensitive recurrent ovarian cancer. We aimed to compare the value of [68Ga]-FAPI and [18F]-FDG PET/CT for the detection of relapsed lesions. Twenty-nine suspected platinum-sensitive recurrent ovarian cancers were enrolled from January 2022 to July 2022. [18F]-FDG and [68Ga]-FAPI PET/CT were obtained within 1week for radiological evaluation. Treatment strategies, visual scores, and Eisenkop scores were recorded. A paired T test was used to compare differences between two scans. Sensitivity, specificity, PPV, NPV, and accuracy were also evaluated. Up to 22 (75.86%) patients displayed inconsistency between two scans. Among them, 4 patients with negative [18F]-FDG imaging had measurable lesions in [68Ga]-FAPI scans. The treatment strategies were changed in 5 patients (17.24%) due to discrepancies. Finally, 15 (35.7%) patients underwent surgeries, and 14 patients achieved complete resection, except for 1 patient who had milliarc residual disease. TBR, but not SUVmax, of [68Ga]-FAPI was significantly higher than that of [18F]-FDG in recurrent lesions. Compared with [18F]-FDG, [68Ga]-FAPI PET presented higher sensitivity and accuracy for lesion detection (96.30% vs. 49.07% and 97.40% vs. 63.87%, respectively). Additionally, [68Ga]-FAPI PET showed a much higher visual score than [18F]-FDG PET (41 vs. 4), especially for peritoneal metastasis (35 vs. 1). [68Ga]-FAPI PET also presented a larger tumor burden than [18F]-FDG according to the Eisenkop score (27 vs. 16, p = 0.025). [68Ga]-FAPI was superior to [18F]-FDG PET/CT for the detection of recurrent lesions, which is pivotal for the individualized management of platinum-sensitive recurrent ovarian cancer.

Single-timepoint low-dimensional characterization and classification of acute versus chronic multiple sclerosis lesions using machine learning

Multiple sclerosis (MS) is a chronic inflammatory and neurodegenerative disease characterized by the appearance of focal lesions across the central nervous system. The discrimination of acute from chronic MS lesions may yield novel biomarkers of inflammatory disease activity which may support patient management in the clinical setting and provide endpoints in clinical trials. On a single timepoint and in the absence of a prior reference scan, existing methods for acute lesion detection rely on the segmentation of hyperintense foci on post-gadolinium T1-weighted magnetic resonance imaging (MRI), which may underestimate recent acute lesion activity. In this paper, we aim to improve the sensitivity of acute MS lesion detection in the single-timepoint setting, by developing a novel machine learning approach for the automatic detection of acute MS lesions, using single-timepoint conventional non-contrast T1- and T2-weighted brain MRI. The MRI input data are supplemented via the use of a convolutional neural network generating “lesion-free” reconstructions from original “lesion-present” scans using image inpainting. A multi-objective statistical ranking module evaluates the relevance of textural radiomic features from the core and periphery of lesion sites, compared within “lesion-free” versus “lesion-present” image pairs. Then, an ensemble classifier is optimized through a recursive loop seeking consensus both in the feature space (via a greedy feature-pruning approach) and in the classifier space (via model selection repeated after each pruning operation). This leads to the identification of a compact textural signature characterizing lesion phenotype. On the patch-level task of acute versus chronic MS lesion classification, our method achieves a balanced accuracy in the range of 74.3–74.6% on fully external validation cohorts.

Improving the detection of new lesions in multiple sclerosis with a cascaded 3D fully convolutional neural network approach.

Longitudinal magnetic resonance imaging (MRI) has an important role in multiple sclerosis (MS) diagnosis and follow-up. Specifically, the presence of new lesions on brain MRI scans is considered a robust predictive biomarker for the disease progression. New lesions are a high-impact prognostic factor to predict evolution to MS or risk of disability accumulation over time. However, the detection of this disease activity is performed visually by comparing the follow-up and baseline scans. Due to the presence of small lesions, misregistration, and high inter-/intra-observer variability, this detection of new lesions is prone to errors. In this direction, one of the last Medical Image Computing and Computer Assisted Intervention (MICCAI) challenges was dealing with this automatic new lesion quantification. The MSSEG-2: MS new lesions segmentation challenge offers an evaluation framework for this new lesion segmentation task with a large database (100 patients, each with two-time points) compiled from the OFSEP (Observatoire français de la sclérose en plaques) cohort, the French MS registry, including 3D T2-w fluid-attenuated inversion recovery (T2-FLAIR) images from different centers and scanners. Apart from a change in centers, MRI scanners, and acquisition protocols, there are more challenges that hinder the automated detection process of new lesions such as the need for large annotated datasets, which may be not easily available, or the fact that new lesions are small areas producing a class imbalance problem that could bias trained models toward the non-lesion class. In this article, we present a novel automated method for new lesion detection of MS patient images. Our approach is based on a cascade of two 3D patch-wise fully convolutional neural networks (FCNNs). The first FCNN is trained to be more sensitive revealing possible candidate new lesion voxels, while the second FCNN is trained to reduce the number of misclassified voxels coming from the first network. 3D T2-FLAIR images from the two-time points were pre-processed and linearly co-registered. Afterward, a fully CNN, where its inputs were only the baseline and follow-up images, was trained to detect new MS lesions. Our approach obtained a mean segmentation dice similarity coefficient of 0.42 with a detection F1-score of 0.5. Compared to the challenge participants, we obtained one of the highest precision scores (PPVL = 0.52), the best PPVL rate (0.53), and a lesion detection sensitivity (SensL of 0.53).

Performance of Aptima HPV E6/E7 mRNA Test for Detection of Cervical Lesions in a Large Chinese Population.

We aimed to examine the effectiveness of Aptima HPV E6/E7 mRNA test for detection of cervical lesions in a large Chinese population. Overall, 4,350 women, who received simultaneously Aptima HPV E6/E7 mRNA test and HPV DNA test, followed by cervical biopsy in the Department of Gynecology of the Second Affiliated Hospital of Soochow University, Jiangsu Province, China from 2016-2020, were recruited. The detection of cervical lesions was compared between Aptima HPV E6/E7 mRNA test and HPV DNA test. Overall, HPV DNA test exhibited a higher detection of all cervical lesions than Aptima HPV E6/E7 mRNA test (P<0.05), and showed a higher efficacy for detection of normal tissues and chronic cervicitis (P<0.05) and low-grade squamous intraepithelial lesions (LSILs) (P<0.05) than Aptima HPV E6/E7 mRNA test; while Aptima HPV E6/E7 mRNA test showed a greater detection of high-grade squamous intraepithelial lesions (HSILs) (P<0.05) and invasive cervical carcinoma than HPV DNA test. Aptima HPV E6/E7 mRNA test exhibited a higher specificity P<0.05), positive and negative prediction rates than HPV DNA test for detection of cervical lesions, and the sensitivity was comparable between the two tests (P>0.05). Aptima HPV E6/E7 mRNA test gradually improves the detection of cervical lesions with disease severity, and shows a higher specificity, positive and negative prediction rates and comparable sensitivity for detection of clinical cervical lesions as compared with HPV DNA test.

NIMG-07. APPLYING A GLIOMA-TRAINED DEEP LEARNING AUTO-SEGMENTATION TOOL ON BM PRE- AND POST-RADIOSURGERY

Abstract PURPOSE Stereotactic radiosurgery (SRS) has become the mainstay to treat BM. Follow-up MRI provides important information on lesion treatment response and guides future therapy planning. Volumetric measurements of BM have shown promise over traditional uni- and two-dimensional measurements in more accurate and repeatable assessment. However, routine clinical use has yet to be achieved because the workflow is laborious. In previous work, we developed a PACS-integrated deep learning algorithm for automatic high- and low-grade glioma 3D segmentation. In this work, we applied this U-Net to segment BM on pre- and post-Gamma Knife (GK) MRI and evaluated the performance. METHODS 10 pre- and post-GK studies were autosegmented in five randomly selected patients (melanoma n= 3, breast n= 2). The glioma trained algorithm segmented the “Whole Tumor” (tumor core+peritumoral edema on T2w-FLAIR) and “Tumor Core” (CE tumor core+necrosis on SPGR). The AI generated segmentation was then revised as needed by a board-certified neuroradiologist and the dice-similarity-coefficient (DSC) between the revised and automatic volumetric segmentations were calculated. RESULTS Four patients had multicentric (2-4 BM) lesions. The mean± SD DSC for Whole Tumor and Tumor Core were 0.92±0.06 and 0.46±0.30 for pretreatment, 0.84±0.09 and 0.41±0.25 for posttreatment BM, respectively. The tool detected lesions with a sensitivity of 45% (5/11) for pretreatment and 50% (3/6) for posttreatment lesions. Three pretreatment and all posttreatment lesions that were not detected by the autosegmentation tool showed a very faint hyperintense peritumoral edema in T2w-FLAIR. CONCLUSION Volumetric segmentation of edema on FLAIR using the glioma-trained segmentation algorithm on pre- and post-GK BM did not require major adjustment of segmentation if it detects the lesion. On the other hand, with low sensitivity of lesion detection and low DSC for enhancing component, dedicated training of the algorithm on annotated BM data will be needed.

Automatic Detection of Periapical Osteolytic Lesions on Cone-beam Computed Tomography Using Deep Convolutional Neuronal Networks

IntroductionCone-beam computed tomography (CBCT) is an essential diagnostic tool in oral radiology. Radiolucent periapical lesions (PALs) represent the most frequent jaw lesions. However, the description, interpretation, and documentation of radiological findings, especially incidental findings, are time-consuming and resource-intensive, requiring a high degree of expertise. To improve quality, dentists may use artificial intelligence in the form of deep learning tools. This study was conducted to develop and validate a deep convolutional neuronal network for the automated detection of osteolytic PALs in CBCT data sets. MethodsCBCT data sets from routine clinical operations (maxilla, mandible, or both) performed from January to October 2020 were retrospectively screened and selected. A 2-step approach was used for automatic PAL detection. First, tooth localization and identification were performed using the SpatialConfiguration-Net based on heatmap regression. Second, binary segmentation of lesions was performed using a modified U-Net architecture. A total of 144 CBCT images were used to train and test the networks. The method was evaluated using the 4-fold cross-validation technique. ResultsThe success detection rate of the tooth localization network ranged between 72.6% and 97.3%, whereas the sensitivity and specificity values of lesion detection were 97.1% and 88.0%, respectively. ConclusionsAlthough PALs showed variations in appearance, size, and shape in the CBCT data set and a high imbalance existed between teeth with and without PALs, the proposed fully automated method provided excellent results compared with related literature.

Artificial intelligence guided enhancement of digital PET: scans as fast as CT?

PurposeBoth digital positron emission tomography (PET) detector technologies and artificial intelligence based image post-reconstruction methods allow to reduce the PET acquisition time while maintaining diagnostic quality. The aim of this study was to acquire ultra-low-count fluorodeoxyglucose (FDG) ExtremePET images on a digital PET/computed tomography (CT) scanner at an acquisition time comparable to a CT scan and to generate synthetic full-dose PET images using an artificial neural network.MethodsThis is a prospective, single-arm, single-center phase I/II imaging study. A total of 587 patients were included. For each patient, a standard and an ultra-low-count FDG PET/CT scan (whole-body acquisition time about 30 s) were acquired. A modified pix2pixHD deep-learning network was trained employing 387 data sets as training and 200 as test cohort. Three models (PET-only and PET/CT with or without group convolution) were compared. Detectability and quantification were evaluated.ResultsThe PET/CT input model with group convolution performed best regarding lesion signal recovery and was selected for detailed evaluation. Synthetic PET images were of high visual image quality; mean absolute lesion SUVmax (maximum standardized uptake value) difference was 1.5. Patient-based sensitivity and specificity for lesion detection were 79% and 100%, respectively. Not-detected lesions were of lower tracer uptake and lesion volume. In a matched-pair comparison, patient-based (lesion-based) detection rate was 89% (78%) for PERCIST (PET response criteria in solid tumors)-measurable and 36% (22%) for non PERCIST-measurable lesions.ConclusionLesion detectability and lesion quantification were promising in the context of extremely fast acquisition times. Possible application scenarios might include re-staging of late-stage cancer patients, in whom assessment of total tumor burden can be of higher relevance than detailed evaluation of small and low-uptake lesions.

Automated Detection of Brain Metastases on T1-Weighted MRI Using a Convolutional Neural Network: Impact of Volume Aware Loss and Sampling Strategy.

Detection of brain metastases (BM) and segmentation for treatment planning could be optimized with machine learning methods. Convolutional neural networks (CNNs) are promising, but their trade-offs between sensitivity and precision frequently lead to missing small lesions. Combining volume aware (VA) loss function and sampling strategy could improve BM detection sensitivity. Retrospective. A total of 530 radiation oncology patients (55% women) were split into a training/validation set (433 patients/1460 BM) and an independent test set (97 patients/296 BM). 1.5 T and 3T, contrast-enhanced three-dimensional (3D) T1-weighted fast gradient echo sequences. Ground truth masks were based on radiotherapy treatment planning contours reviewed by experts. A U-Net inspired model was trained. Three loss functions (Dice, Dice + boundary, and VA) and two sampling methods (label and VA) were compared. Results were reported with Dice scores, volumetric error, lesion detection sensitivity, and precision. A detected voxel within the ground truth constituted a true positive. McNemar's exact test to compare detected lesions between models. Pearson's correlation coefficient and Bland-Altman analysis to compare volume agreement between predicted and ground truth volumes. Statistical significance was set at P ≤ 0.05. Combining VA loss and VA sampling performed best with an overall sensitivity of 91% and precision of 81%. For BM in the 2.5-6 mm estimated sphere diameter range, VA loss reduced false negatives by 58% and VA sampling reduced it further by 30%. In the same range, the boundary loss achieved the highest precision at 81%, but a low sensitivity (24%) and a 31% Dice loss. Considering BM size in the loss and sampling function of CNN may increase the detection sensitivity regarding small BM. Our pipeline relying on a single contrast-enhanced T1-weighted MRI sequence could reach a detection sensitivity of 91%, with an average of only 0.66 false positives per scan. 3 TECHNICAL EFFICACY: Stage 2.

Accuracy of blood-based biomarkers for screening precancerous colorectal lesions: a protocol for systematic review and meta-analysis

IntroductionColorectal cancer (CRC) is the third most diagnosed cancer and the second most common cause of cancer mortality worldwide. Most CRCs develop through either the adenoma-to-carcinoma or the serrated pathways,...

Abstract P3-02-11: Towards a new standard for staging: A comparative study of [18F]FES PET/CT vs [18F]FDG PET/CT in patients with clinical stage II/III and locoregional recurrent estrogen receptor positive breast cancer

Abstract Introduction: Accurate staging of patients with locally advanced breast cancer or locoregional recurrent (LRR) breast cancer is essential for providing a tailored treatment. According to current guidelines, staging is preferably done with positron emission tomography (PET) using 2-[18F]fluoro-2-deoxy-D-glucose ([18F]FDG), combined with diagnostic computed tomography (CT). Previous studies suggest that [18F]FDG PET might underperform in patients with grade 1-2, estrogen receptor positive (ER+) breast cancer as these tumors are often metabolically less active than other subtypes of breast cancer. Thus, there is a need to improve current diagnostics for this population. 16α-[18F]-fluoro-17β-estradiol ([18F]FES), radioactive labeled estradiol, has emerged as a powerful imaging tool for in vivo visualization of ER+ lesions. The aim of this study was to investigate whether [18F]FES PET/CT improves staging of patients with clinical stage II/III or LRR ER+ breast cancer, compared to [18F]FDG PET/CT. Methods: Patients with clinical stage II/III or LRR 1-2, ER+ breast cancer were included in this prospective multicenter clinical trial (NCT03726931). All patients underwent a [18F]FES PET/CT and a [18F]FDG PET/CT scan. Both PET scans were visually assessed, independently of each other. Visually suspect lesions for malignancy were verified pathologically (by biopsy or fine needle aspiration) or by conventional imaging (mammography, ultrasound, magnetic resonance imaging, CT). The stage of the disease was determined twice in a multidisciplinary team, based on conventional diagnostics (conventional imaging and pathology) and [18F]FDG PET/CT or [18F]FES PET/CT, respectively. Semi-quantitative PET parameters, standardized uptake values (SUVs) and tumor-to-blood ratios (TBRs), were obtained by defining volumes of interest in the aorta ascendens and in suspect tumor lesions. These PET parameters were correlated with various pathological features, such as histological subtype, grade and ER, progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2) expression. Results: A total of 40 patients were included. On patient level, the sensitivity of [18F]FES PET/CT and [18F]FDG PET/CT to detect ER+ disease was 87.5% and 82.5%, respectively. [18F]FES PET/CT correctly determined the stage in 34/40 patients (85.0%) whereas [18F]FDG PET/CT correctly determined the stage in 29/40 patients (72.5%). [18F]FES PET/CT correctly changed the stage in 8/40 patients (20.0%) with upstaging in 6 patients and downstaging in 2 patients, compared to [18F]FDG PET/CT. On lesion level, a total of 140 suspect lesions were analyzed: 82 lesions were pathologically verified whereas 58 were radiographically verified. The sensitivity for lesion detection on [18F]FES PET/CT and [18F]FDG PET/CT was 90.6% and 77.3%, respectively. [18F]FES PET/CT identified more lesions correctly: 120/140 (85.7%) lesions vs 103/140 (73.6%) on [18F]FES PET/CT vs [18F]FDG PET/CT, respectively. [18F]FES PET/CT also missed less lesions (12/140, 8.6%) compared to [18F]FDG PET/CT (29/140, 20.7%). [18F]FES SUVmax, SUVmean, SUVpeak and TBR correlated with ductal subtype and percentage of PR expression (p &amp;lt;0.05, positive correlation). Conclusion: [18F]FES PET/CT improves disease staging in patients with stage II/III and LRR grade 1-2, ER+ breast cancer, compared to [18F]FDG PET/CT. This is in line with the ability of the tracer to identify malignant lesions more accurately in this patient population and supports further development of [18F]FES for staging grade 1-2, ER+ breast cancer. Citation Format: Ramsha Iqbal, Emma C Bonjer, Annette van Zweeden, Lemonitsa H Mammatas, Jaap JM Teunissen, Jan Booij, Gwen MP Diepenhorst, Laura Schijf, Susanne van der Velde, Katya Duvivier, Ellis Barbé, Daniela E Oprea-Lager, Catharina W Menke-van der Houven van Oordt. Towards a new standard for staging: A comparative study of [18F]FES PET/CT vs [18F]FDG PET/CT in patients with clinical stage II/III and locoregional recurrent estrogen receptor positive breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-02-11.

Diagnostic accuracy of double inversion recovery (DIR) in detection of cortical gray matter lesions in patients with MS

BackgroundConventional imaging techniques have a low sensitivity for detection of cortical and deep grey matter lesions in MS which hinder accurate assessment of the total lesion burden. Aim of this work was to assess the diagnostic accuracy of double inversion recovery (DIR) sequence in the detection of cortical grey matter lesions in MS patients.ResultsForty MS patients were prospectively included in this study. Imaging was performed using Philips Ingenia 1.5 T device. The sensitivity, specificity, PPV, NPV and accuracy of DIR sequence in detection of cortical grey matter lesions were 60%, 100%, 100%, 55.6% and 73.3%, respectively. The sensitivity, specificity, positive and negative predictive values as well as accuracy of Flair sequence were 50%, 100%, 100%, 50% and 66.7%, respectively. The sensitivity, specificity, positive and negative predictive values as well as accuracy of T2 sequence in the detection of cortical grey matter lesions were 22.5%, 100%, 100%, 39.2% and 48.3%, respectively.ConclusionsDetection rate of cortical gray matter lesions was significantly higher on DIR sequence than on T2 and Flair sequences.