Multiple Scans Research Articles

A deep learning model for generating [18F]FDG PET Images from early-phase [18F]Florbetapir and [18F]Flutemetamol PET images

IntroductionAmyloid-β (Aβ) plaques is a significant hallmark of Alzheimer's disease (AD), detectable via amyloid-PET imaging. The Fluorine-18-Fluorodeoxyglucose ([18F]FDG) PET scan tracks cerebral glucose metabolism, correlated with synaptic dysfunction and disease progression and is complementary for AD diagnosis. Dual-scan acquisitions of amyloid PET allows the possibility to use early-phase amyloid-PET as a biomarker for neurodegeneration, proven to have a good correlation to [18F]FDG PET. The aim of this study was to evaluate the added value of synthesizing the later from the former through deep learning (DL), aiming at reducing the number of PET scans, radiation dose, and discomfort to patients.MethodsA total of 166 subjects including cognitively unimpaired individuals (N = 72), subjects with mild cognitive impairment (N = 73) and dementia (N = 21) were included in this study. All underwent T1-weighted MRI, dual-phase amyloid PET scans using either Fluorine-18 Florbetapir ([18F]FBP) or Fluorine-18 Flutemetamol ([18F]FMM), and an [18F]FDG PET scan. Two transformer-based DL models called SwinUNETR were trained separately to synthesize the [18F]FDG from early phase [18F]FBP and [18F]FMM (eFBP/eFMM). A clinical similarity score (1: no similarity to 3: similar) was assessed to compare the imaging information obtained by synthesized [18F]FDG as well as eFBP/eFMM to actual [18F]FDG. Quantitative evaluations include region wise correlation and single-subject voxel-wise analyses in comparison with a reference [18F]FDG PET healthy control database. Dice coefficients were calculated to quantify the whole-brain spatial overlap between hypometabolic ([18F]FDG PET) and hypoperfused (eFBP/eFMM) binary maps at the single-subject level as well as between [18F]FDG PET and synthetic [18F]FDG PET hypometabolic binary maps.ResultsThe clinical evaluation showed that, in comparison to eFBP/eFMM (average of clinical similarity score (CSS) = 1.53), the synthetic [18F]FDG images are quite similar to the actual [18F]FDG images (average of CSS = 2.7) in terms of preserving clinically relevant uptake patterns. The single-subject voxel-wise analyses showed that at the group level, the Dice scores improved by around 13% and 5% when using the DL approach for eFBP and eFMM, respectively. The correlation analysis results indicated a relatively strong correlation between eFBP/eFMM and [18F]FDG (eFBP: slope = 0.77, R2 = 0.61, P-value < 0.0001); eFMM: slope = 0.77, R2 = 0.61, P-value < 0.0001). This correlation improved for synthetic [18F]FDG (synthetic [18F]FDG generated from eFBP (slope = 1.00, R2 = 0.68, P-value < 0.0001), eFMM (slope = 0.93, R2 = 0.72, P-value < 0.0001)).ConclusionWe proposed a DL model for generating the [18F]FDG from eFBP/eFMM PET images. This method may be used as an alternative for multiple radiotracer scanning in research and clinical settings allowing to adopt the currently validated [18F]FDG PET normal reference databases for data analysis.

Enhanced triboelectricity through visible-light-induced surface charges in BTO-polymer hybrid for coexistence solar-mechanical energy harvesting

The exploration of hybrid composites holds great promise in the pursuit of synergistic energy-harvesting solutions, providing an efficient approach to tap into multiple energy sources. One striking example is the BTO (BaTiO3)-polymer hybrid where its high dielectric constant and the piezo-/ferroelectricity are leveraged to improve the triboelectricity of the polymer-based triboelectric nanogenerator (TENG). Beyond this, the BTO also exhibits a photoactive nature, which, until now, has not been exploited to enhance triboelectricity. In this study, a facile method to exploit BTO’s photoresponses in a BTO-polymer hybrid is reported, where surface states present in BTO nanoparticles enable visible spectrum absorption, and the interfaces are designed to facilitate charge spatial separation. Upon visible light illumination, surface charges are generated on the BTO-polymer hybrid, significantly enhancing the photo-induced charge electrification, which in turn boosts the TENG output. These findings demonstrate the possibility of simultaneously harvesting solar and mechanical energies in TENGs using ceramic-polymer hybrids. Additionally, the study employs multiple advanced Scanning Probe Microscopy (SPM) techniques to elucidate the roles of each component and interface in energy harvesting, shedding light on the functional material design. This work not only broadens the variety of energy sources for TENGs but also addresses the growing demand for sustainable and adaptable methods of power generation.

Variational model-based reconstruction techniques for multi-patch data in Magnetic Particle Imaging

Magnetic Particle Imaging is an emerging imaging modality through which it is possible to detect tracers containing superparamagnetic nanoparticles. The exposure of the particles to dynamic magnetic fields generates a non-linear response that is used to locate the particles and produce an image of their distribution. The bounding box that can be covered by a single scan curve depends on the strength of the gradients of the magnetic fields applied, which is limited due to the risk of causing peripheral nerve stimulation (PNS) in the patients. To address this issue, multiple scans are performed. The scan data must be merged together to produce reconstructions of larger regions of interest. In this paper we propose a mathematical framework which can deal with rather general multi-patching scenarios including rigid transformations of the field of view (FoV), the specimen and of the scanner. We show the flexibility of this framework in a variety of different scanning scenarios. Moreover, we describe an iterative reconstruction algorithm that yields a reconstruction of the target distribution by minimizing a convex functional which includes positivity constraints and sparsity enforcing priors. We show its convergence to a minimizer and perform numerical experiments on simulated data.

Towards interactive AI-authoring with prototypical few-shot classifiers in histopathology

A vast multitude of tasks in histopathology could potentially benefit from the support of artificial intelligence (AI). Many examples have been shown in the literature and first commercial products with FDA or CE-IVDR clearance are available. However, two key challenges remain: (1) a scarcity of thoroughly annotated images, respectively the laboriousness of this task, and (2) the creation of robust models that can cope with the data heterogeneity in the field (domain generalization). In this work, we investigate how the combination of prototypical few-shot classification models and data augmentation can address both of these challenges. Based on annotated data sets that include multiple centers, multiple scanners, and two tumor entities, we examine the robustness and the adaptability of few-shot classifiers in multiple scenarios. We demonstrate that data from one scanner and one site are sufficient to train robust few-shot classification models by applying domain-specific data augmentation. The models achieved classification performance of around 90% on a multiscanner and multicenter database, which is on par with the accuracy achieved on the primary single-center single-scanner data. Various convolutional neural network (CNN) architectures can be used for feature extraction in the few-shot model. A comparison of nine state-of-the-art architectures yielded that EfficientNet B0 provides the best trade-off between accuracy and inference time. The classification of prototypical few-shot models directly relies on class prototypes derived from example images of each class. Therefore, we investigated the influence of prototypes originating from images from different scanners and evaluated their performance also on the multiscanner database. Again, our few-shot model showed a stable performance with an average absolute deviation in accuracy compared to the primary prototypes of 1.8% points. Finally, we examined the adaptability to a new tumor entity: classification of tissue sections containing urothelial carcinoma into normal, tumor, and necrotic regions. Only three annotations per subclass (e.g., muscle and adipose tissue are subclasses of normal tissue) were provided to adapt the few-shot model, which obtained an overall accuracy of 93.6%. These results demonstrate that prototypical few-shot classification is an ideal technology for realizing an interactive AI authoring system as it only requires few annotations and can be adapted to new tasks without involving retraining of the underlying feature extraction CNN, which would in turn require a selection of hyper-parameters based on data science expert knowledge. Similarly, it can be regarded as a guided annotation system. To this end, we realized a workflow and user interface that targets non-technical users.

Place Recognition through Multiple LiDAR Scans Based on the Hidden Markov Model.

Autonomous driving systems for unmanned ground vehicles (UGV) operating in enclosed environments strongly rely on LiDAR localization with a prior map. Precise initial pose estimation is critical during system startup or when tracking is lost, ensuring safe UGV operation. Existing LiDAR-based place recognition methods often suffer from reduced accuracy due to only matching descriptors from individual LiDAR keyframes. This paper proposes a multi-frame descriptor-matching approach based on the hidden Markov model (HMM) to address this issue. This method enhances the place recognition accuracy and robustness by leveraging information from multiple frames. Experimental results from the KITTI dataset demonstrate that the proposed method significantly enhances the place recognition performance compared with the scan context-based single-frame descriptor-matching approach, with an average performance improvement of 5.8% and with a maximum improvement of 15.3%.

Reducing Metal Artifacts in X-ray Computed Tomography Based on Metal Effect Estimation and Multiple CT Volume Fusion

In recent years, X-ray Computed Tomography (CT) has been widely used for investigating industrial parts. Many industrial parts often include metals, which cause various CT artifacts such as beam hardening artifacts, and metal artifacts. These CT artifacts lower the quality of X-ray CT images, prevent precise measurements. In order to reduce metal artifacts, here we propose using multiple X-ray CT scanning. Our multiple X-ray CT scanning method includes local artifacts estimation, proposing adequate multiple scanning angles, multiple scanned data registration and multiple CT data fusion. Based on local artifacts estimation, optimal multiple scan angles for data fusion are selected. Data registration and fusion are also based on local artifacts estimation, excluding effects of artifacts during the process. In order to demonstrate our method, we measured industrial parts which includes metals and resins, proposed optimal multiple scanning angles and conducted volume fusion. As a result, the metal artifacts in the CT volume reduced significantly compared to single X-ray CT scanning. Experiments also showed that metal artifacts in different regions are reduced by additional scans, which suggest that our scan angle selection method successfully selected optimal multiple scan angles.

Comparison of clinical outcomes of stable disease with confirmed tumor reduction and RECIST partial response for tebentafusp in metastatic uveal melanoma (mUM).

9529 Background: Tebentafusp (gp100 x CD3) has demonstrated an overall survival (OS) benefit in mUM. Benefit was observed in patients (pts) who achieved a RECIST v1.1 partial response (PR), stable disease (SD) and even progressive disease (PD) (Nathan 2021). In Phase (Ph) 1, some melanoma pts had SD with tumor reduction &gt; 10% that was confirmed at ≥ 1 subsequent scan (referred to as minor response, MR). In Ph 2, an analysis of MR was pre-specified as an endpoint and presented here. Methods: 127 HLA-A*02:01+ pts with previously treated mUM received weekly intravenous tebentafusp following intra-pt dose escalation of 20mcg Week 1, 30mcg Week 2 and 68mcg Week 3+ (NCT02570308; Carvajal 2022). Radiologic assessments were performed every 8 weeks until week 40, then q12 weeks. Tumor assessment was evaluated by a blinded independent review committee per RECISTv1.1. MR was prospectively defined as RECISTv1.1 best response of SD with reduction in sum of target lesions of -10% to -29% and which was confirmed ≥ 4 weeks later. ctDNA levels were assessed using a targeted mPCR-NGS assay for mutations in 15 genes including mUM oncogenes GNAQ, GNA11, SF3B1, CYSLTR2, PLCB4 and EIF1AX (Natera). Molecular response was defined as ≥ 0.5 log [68%] ctDNA reduction by week 9. Data cut off: Oct 2022. Median duration of follow-up was 46 months. Results: Of 127 pts, the clinical benefit rate of PR + SD was 50% (64/127). 25% (32/127) had any tumor reduction that was confirmed on ≥ 1 subsequent scan, including 6 PR (ORR 5%) and 26 SD (20%). 8/26 SD (6% of 127) met the pre-defined threshold for MR, most (5/8) had &gt;20% reduction. The median duration of response for PR and MR were 8.7 months and 10.6 months, respectively. The estimated percent of pts with PR and MR remaining in response at Month 20 were 20% and 33%, respectively. 3/6 PR pts were alive ≥ 3 years vs 3/8. 58% of PR + SD pts with evaluable ctDNA had a molecular response (26/45), including 2/4 PRs and 5/5 MRs. Conclusions: This Ph2 UM study prospectively confirmed that a subset of SD patients had tumor reduction over multiple scans. The frequency of minor response was similar to that of PR and had similar durability, OS and ctDNA molecular response. SD with confirmed tumor reduction is an emerging endpoint for the ImmTAC platform and will be studied in other trials (NCT05549297, NCT04262466). Clinical trial information: NCT02570308 .

Microstructure Tailoring for High Strength Ti-6Al-4V without Alloying Elements through Optimized Preheating and Post-Heating Laser Scanning in Laser Powder Bed Fusion

Ti-6Al-4V with its eclectic array of excellent properties along with the combination of meticulous precision and flexibility offered by the laser powder bed fusion (LPBF) technology makes it a strong proponent in the field of engineering applications. As a substantial amount of research has paved the way to fabricate Ti-6AL-4V more effectively and efficiently, researchers are becoming more adventurous in finding out the optimal techniques to get better yields in terms of mechanical responses. This includes post-processing techniques i.e., heat treatment (HT) or introducing various alloying elements. Nevertheless, these techniques not only make the overall fabrication more expensive and time-consuming but also contradict the simplistic notion of additive manufacturing (AM) by imparting multistage fabrication without a considerable improvement overall. Here, we propose an innovative breakthrough in the field of Ti-6AL-4V fabrication with LPBF by introducing an in-situ approach to tackle the handicap mentioned in contemporary studies. By imparting multiple laser scans prior to and after the melting scan at each layer, a remarkable 37% improvement in yield strength (YS) can be achieved with higher elongation, while also maintaining a high relative density of around 99.99%.

Optimization of sample preparation, fluorescence- and Raman techniques for environmental microplastics

Microspectroscopic imaging techniques based on spontaneous Raman scattering, Stimulated Raman Scattering (SRS), or fluorescence (with a selective dye) can be used to detect environmental microplastics (MPs) and determine their chemical as well as physical properties. The present study first focuses on optimizing the sample preparation, including a new design for a density separation apparatus and optimization of the Nile Red staining procedure. Tests were carried out with both white and colored reference materials, as well as environmental MPs in a suspended matter sample from the Rhine river. The new ‘MESSY’ system has a mean recovery of 95 ± 5.5 % (three polymer materials, in duplicate). The optimized Nile Red staining allows coarse categorization of MPs into “polar” vs. “non-polar” materials based on their Fluorescence Index (emission wavelength), but fluorescent additives in the polymer can cause misclassification. For unambiguous identification of the polymer type, Raman spectroscopy can be used. Even colored polymers, with or without Nile Red staining, were readily identified by Raman spectroscopy using a red laser (785 nm), except for particles containing carbon black. A Deep-UV Raman microscope (ex = 248.6 nm) was constructed, which allowed identification of all colored plastics, even those pigmented with carbon black. Since unsupervised mapping with spontaneous Raman is very slow, point measurements are preferably used after preselection of particles of interest based on fluorescence imaging. SRS is several orders of magnitude faster than spontaneous Raman mapping but requires multiple scans at different z-heights and at multiple wavenumber settings to detect and identify all particles. The results are expected to contribute to the development of suitable methodologies for the detection and identification of environmental microplastics.

In-fiber illumination-detection-synchronized confocal microscopy for high-precision fast axial scanning

Confocal microscopy remains a major workhorse in the 3D surface topography measurement owing to its reliability and compatibility for workpieces with various reflectance, which, however, suffers from slow scanning speed and limited axial resolution. Commercial confocal microscopy primarily relies on the axial movement of a sample or objective lens to create a series of 2D image stacks, and then reconstruct 3D map. Nevertheless, most samples and objective lens are too bulky to scan rapidly, let alone that multiple scans are required to increase the signal-to-noise ratio (SNR). To address this issue, an original approach named illumination-detection-synchronized confocal microscopy was developed, which enabled high-precision fast axial scanning via a compact fiber-multiplexed port with the high-speed synchronous motion of point source and point detection. The approach can effectively improve the peak localization precision while not requiring a high-precision motion stage. Theoretically, via controlling the magnification of the optical system M > 1, the total peak localization precision of axial light intensity response curve can be improved by M3 times. Experiments show that the single-point repeat measurement precision by this approach is improved by 4 times, and the profile measurement precision is enhanced by 2.3 times. Moreover, this approach can effectively avoid collisions between the sample and objective lens, and has the potential to design a low-cost single-point confocal sensor or confocal profilometer.

Precision brain morphometry using cluster scanning

Abstract Measurement error limits the statistical power to detect group differences and longitudinal change in structural MRI morphometric measures (e.g., hippocampal volume, prefrontal cortical thickness). Recent advances in scan acceleration enable extremely fast T1-weighted scans (~1 minute) that achieve morphometric errors that are close to the errors in longer traditional scans. As acceleration allows multiple scans to be acquired in rapid succession, it becomes possible to pool estimates to increase measurement precision, a strategy known as “cluster scanning.” Here, we explored brain morphometry using cluster scanning in a test-retest study of 40 individuals (12 younger adults, 18 cognitively unimpaired older adults, and 10 adults diagnosed with mild cognitive impairment or Alzheimer’s Dementia). Morphometric errors from a single compressed sensing (CS) 1.0 mm scan (CS) were, on average, 12% larger than a traditional scan using the Alzheimer’s Disease Neuroimaging Initiative (ADNI) protocol. Pooled estimates from four clustered CS acquisitions led to errors that were 34% smaller than ADNI despite having a shorter total acquisition time. Given a fixed amount of time, a gain in measurement precision can thus be achieved by acquiring multiple rapid scans instead of a single traditional scan. Errors were further reduced when estimates were pooled from eight CS scans (51% smaller than ADNI). Neither pooling across a break nor pooling across multiple scans of different spatial resolutions boosted this benefit. We discuss the potential of cluster scanning to improve morphometric precision, boost statistical power, and produce more sensitive disease progression biomarkers.

Determination of Triacylglycerol Composition in Mealworm Oil (Tenebrio molitor) via Electrospray Ionization Tandem Mass Spectrometry with Multiple Neutral Loss Scans.

Mealworms (Tenebrio molitor) have been used as an alternative source of proteins and lipids. Triacylglycerols (TAGs) are major sources of energy and have been used to provide essential fatty acids. They are also the main components of mealworm oil, and their composition and content are extensively linked to its physical and chemical properties. However, because of the complexity of TAG molecules, their identification and quantitation are challenging. This study employed electrospray ionization tandem mass spectrometry (ESI-MS/MS) with multiple neutral loss scans (NLS) to analyze the TAG composition and content in mealworm oil. Identifying and quantifying TAGs using ESI-MS/MS in combination with multiple NLS was an efficient way to improve accuracy and timeliness. For the accurate quantification of TAGs, isotopic deconvolution and correlation factors were applied. A total of 57 TAGs were identified and quantified: C52:2 (16:0/18:1/18:1) (1549.4 nmol/g, 18.20%), C52:3 (16:0/18:1/18:2) (1488.1 nmol/g, 17.48%), C54:4 (18:1/18:1/18:2) (870.1 nmol/g, 10.23%), C54:6 (18:1/18:2/18:2) (659.8 nmol/g, 7.76%) and C52:4 (16:0/18:2/18:2) (600.5 nmol/g, 7.06%), which were the most abundant TAGs present in the mealworm oil. The fundamental properties of mealworm oil, including its degree of oxidation, nutritional effect and physical properties, were elucidated.

Effects of Laser Scanning Strategy on Bending Behavior and Microstructure of DP980 Steel

Laser bending is a kind of cumulative forming technology and bending efficiency is one of its most important indexes. This study investigates the bending behavior and the microstructure of DP980 steel plates under different laser scanning strategies, using an IPG laser system. Two sets of experiments varied the accumulated line energy density (AED) by altering the laser scanning velocity and number of scans. The results show that, for the single laser scanning process, the bending angle of the plate increases with AED, due to a larger temperature gradient through the thickness direction; however, this relationship is nonlinear. A higher AED led to a sharper initial increase in bending angle, which then plateaued. Under the same AED conditions, the bending angle of the plate undergoing multiple laser scans increases by at least 26% compared to the single one, due to the microstructure changes. It is revealed that the bending efficiency is affected by both the AED and the resultant microstructure evolution in the DP980 steel. Higher AED values and appropriate peak temperatures facilitate better bending behavior due to the formation of uniform martensite and grain refinement. Conversely, excessive peak temperatures can hinder bending due to grain growth.

Cranial Computer Tomography with Photon Counting and Energy-Integrated Detectors: Objective Comparison in the Same Patients.

This study provides an objective comparison of cranial computed tomography (CT) imaging quality and radiation dose between photon counting detectors (PCCTs) and energy-integrated detectors (EIDs). We retrospectively analyzed 158 CT scans from 76 patients, employing both detector types on the same individuals to ensure a consistent comparison. Our analysis focused on the Computed Tomography Dose Index and the Dose-Length Product together with the contrast-to-noise ratio and the signal-to-noise ratio for brain gray and white matter. We utilized standardized imaging protocols and consistent patient positioning to minimize variables. PCCT showed a potential for higher image quality and lower radiation doses, as highlighted by this study, thus achieving diagnostic clarity with reduced radiation exposure, underlining its significance in patient care, particularly for patients requiring multiple scans. The results demonstrated that while both systems were effective, PCCT offered enhanced imaging and patient safety in neuroradiological evaluations.

Explainable Multimodal Graph Isomorphism Network for Interpreting Sex Differences in Adolescent Neurodevelopment

Explainable Multimodal Graph Isomorphism Network for Interpreting Sex Differences in Adolescent Neurodevelopment

A NEW INSTRUMENT TO SUPPORT EXAMINATION OF DISTAL RADIUS ON WRIST RADIOGRAPH

Objectives: To construct an instrument that supports wrist position control according to a number of shooting angles in anteroposterior and lateral position during X-ray examination. Materials and methods: Term such as criteria for evaluating wrist X-ray films, current technical status, some concepts about posture have been scientifically researched. Based on those things to provide research directions and design tool samples. A male volunteer, 36 years old agree to participate in the study, X-rays were taken in an anteroposterior and lateral position for investigation. Results and Conclusion: Initial evaluation shows that this tool is easy to use and only takes a little time to get used to the adjustment operations. The benefit that the device brings is that it helps stabilize the wrist position during X-rays, helping to adjust the angles within the range of motion of a normal person, produces more accurate wrist X-ray results, without having to take multiple scans, saving effort, machine wear and tear costs, creating practical benefits for patients, medical staff, researchers and hospital.

Alternating low-rank tensor reconstruction for improved multiparametric mapping with cardiovascular MR Multitasking.

To develop a novel low-rank tensor reconstruction approach leveraging the complete acquired data set to improve precision and repeatability of multiparametric mapping within the cardiovascular MR Multitasking framework. A novel approach that alternated between estimation of temporal components and spatial components using the entire data set acquired (i.e., including navigator data and imaging data) was developed to improve reconstruction. The precision and repeatability of the proposed approach were evaluated on numerical simulations, 10 healthy subjects, and 10 cardiomyopathy patients at multiple scan times for 2D myocardial T1/T2 mapping with MR Multitasking and were compared with those of the previous navigator-derived fixed-basis approach. In numerical simulations, the proposed approach outperformed the previous fixed-basis approach with lower T1 and T2 error against the ground truth at all scan times studied and showed better motion fidelity. In human subjects, the proposed approach showed no significantly different sharpness or T1/T2 measurement and significantly improved T1 precision by 20%-25%, T2 precision by 10%-15%, T1 repeatability by about 30%, and T2 repeatability by 25%-35% at 90-s and 50-s scan times The proposed approach at the 50-s scan time also showed comparable results with that of the previous fixed-basis approach at the 90-s scan time. The proposed approach improved precision and repeatability for quantitative imaging with MR Multitasking while maintaining comparable motion fidelity, T1/T2 measurement, and septum sharpness and had the potential for further reducing scan time from 90 s to 50 s.

Animal PET scanner with a large field of view is suitable for high-throughput scanning of rodents.

In preclinical studies, high-throughput positron emission tomography (PET) imaging, known as simultaneous multiple animal scanning, can reduce the time spent on animal experiments, the cost of PET tracers, and the risk of synthesis of PET tracers. It is well known that the image quality acquired by high-throughput imaging depends on the PET system. Herein, we investigated the influence of large field of view (FOV) PET scanner on high-throughput imaging. We investigated the influence of scanning four objects using a small animal PET scanner with a large FOV. We compared the image quality acquired by four objects scanned with the one acquired by one object scanned using phantoms and animals. We assessed the image quality with uniformity, recovery coefficient (RC), and spillover ratio (SOR), which are indicators of image noise, spatial resolution, and quantitative precision, respectively. For the phantom study, we used the NEMA NU 4-2008 image quality phantom and evaluated uniformity, RC, and SOR, and for the animal study, we used Wistar rats and evaluated the spillover in the heart and kidney. In the phantom study, four phantoms had little effect on imaging quality, especially SOR compared with that for one phantom. In the animal study as well, four rats had little effect on spillover from the heart muscle and kidney cortex compared with that for one rat. This study demonstrated that an animal PET scanner with a large FOV was suitable for high-throughput imaging. Thus, the large FOV PET scanner can support drug discovery and bridging research through rapid pharmacological and pathological evaluation.

Abstract PO2-14-12: Accurate quantification of slide-level HER2 scores in breast cancer using a machine-learning model, AIM-HER2 Breast Cancer

Abstract Background: HER2 expression level is a key factor in determining the optimal treatment course for breast cancer patients. Roughly 15% of breast cancers are HER2(+), and determination of HER2 status is routinely assessed by immunohistochemistry (IHC). Accurate assessment of the HER2 IHC score (0, 1+, 2+, 3+) by pathologists is therefore critical, especially in light of novel therapeutic approaches demonstrating efficacy in the HER2-low setting (IHC scores 1+, and 2+/FISH-)1,2. To assist pathologists with the consistent provision of reproducible and accurate scores across the entire HER2 scoring range, we developed a machine-learning model (“AIM-HER2”) to generate accurate, slide-level HER2 scores aligned with ASCO-CAP guidelines in clinical breast cancer HER2 IHC specimens. Methods: AIM-HER2 was developed using whole-slide images (WSI; N=4261) from clinical and commercial sources. WSI were split into training (N=2694, 63%) and optimization (N=1567, 37%) sets. An additive multiple instance learning (aMIL) model3 was trained to predict HER2 scores directly from WSI and create interpretable heatmaps that depict HER2 predictions in tissue images. Image artifacts and in situ carcinomas were identified using previously trained artifact and tissue segmentation models and were excluded, leaving only regions of invasive carcinoma to be analyzed. AIM-HER2 performance was assessed on additional slides obtained from five academic or commercial sources (N=804 total, 770 evaluable) on which HER2 IHC was performed. Board-certified pathologists (N=52) with relevant experience provided manual HER2 scores based on ASCO-CAP guidelines. Nested pairwise non-inferiority analysis4 was used to compare model performance to that of pathologists (N=3 pathologists per slide). In the nested pairwise framework, agreement among pathologists was compared to agreement between AIM-HER2 and pathologists via linear kappa, so that summary metrics account for inter-pathologist variability. Results: High concordance was observed between AIM-HER2-predicted and pathologist-labeled slide-level HER2 scores, both overall and for each scoring level. Similar results were observed when assessing AIM-HER2 performance on multiple slide scanners and after IHC with multiple HER2 IHC antibody clones. Results are summarized in Table 1. Conclusions: We developed AIM-HER2, a novel aMIL-based approach for predicting slide-level HER2 IHC scores. AIM-HER2 has similar levels of agreement with pathologists as pathologists have with each other for determining HER2 score. This result is upheld when slides imaged using multiple scanning platforms and stained using multiple HER2 antibody clones. The performance of AIM-HER2 on multiple scanners and after multiple assays supports broad applicability of this algorithm in clinical laboratories, including for the identification of HER2-low cases. Work is ongoing to perform similar analyses in an independent, real-world dataset.

SAN: Mitigating spatial covariance heterogeneity in cortical thickness data collected from multiple scanners or sites.

In neuroimaging studies, combining data collected from multiple study sites or scanners is becoming common to increase the reproducibility of scientific discoveries. At the same time, unwanted variations arise by using different scanners (inter-scanner biases), which need to be corrected before downstream analyses to facilitate replicable research and prevent spurious findings. While statistical harmonization methods such as ComBat have become popular in mitigating inter-scanner biases in neuroimaging, recent methodological advances have shown that harmonizing heterogeneous covariances results in higher data quality. In vertex-level cortical thickness data, heterogeneity in spatial autocorrelation is a critical factor that affects covariance heterogeneity. Our work proposes a new statistical harmonization method called spatial autocorrelation normalization (SAN) that preserves homogeneous covariance vertex-level cortical thickness data across different scanners. We use an explicit Gaussian process to characterize scanner-invariant and scanner-specific variations to reconstruct spatially homogeneous data across scanners. SAN is computationally feasible, and it easily allows the integration of existing harmonization methods. We demonstrate the utility of the proposed method using cortical thickness data from the Social Processes Initiative in the Neurobiology of the Schizophrenia(s) (SPINS) study. SAN is publicly available as an R package.