Computational Platform Research Articles

Integration of a machine learning model-generated surgical risk score into the EHR and preoperative workflow.

11163 Background: We previously developed a machine learning model (MLM) that accurately predicts postoperative complications (POC) for cancer inpatients. As laborious manual entry of patient data is a barrier for risk calculator use, and risk constantly changes depending on clinical status, we sought to integrate our MLM into the electronic health record (EHR) and preoperative workflow as a readily available surgical risk score (SRS) that is generated using real-time data. Here, we report on the integration process and survey results regarding its functionality and impact on patient care. Methods: A MLM that predicts severe POC (CD Grade ≥3) for cancer inpatients undergoing same-hospitalization operations was previously developed using EHR data. To integrate our MLM into the Epic EHR, we developed a real-time infrastructure (RTI) using the cloud-based platform Azure. EHR data is inputted to RTI and processed in real-time to generate model outputs. These outputs are then sent back to the EHR through the Epic Cognitive Computing Platform to generate each patient’s specific SRS. The SRS is dichotomized as “high” or “low” risk based upon a previously determined threshold and is displayed as a “Best Practice Advisory (BPA)” when a surgical case request is entered. In addition to the automated BPA, the SRS is sent to the primary surgeon as an InBasket Case Message. To elicit feedback of our EHR-integrated SRS, a survey was distributed to surgical providers at our institution. Results: After integration of our MLM into the EHR preoperative workflow, 185 surgeries were completed between November 16, 2021 to July 31, 2023. The 30-day severe POC rate was 27.6%, and our MLM had precision of 44.7% and recall of 74.5%, resulting in AUROC of 0.78 and AUPRC of 0.53. 35 surgical providers (40% primary surgeons, 34% surgical fellows, 26% advanced practice providers) from the divisions of surgical oncology, colorectal surgery, and gynecologic oncology completed our survey in July 2022 and July 2023. 28 (80%) of providers stated that they have interacted with the SRS through either BPA or InBasket Case Messages. Of these providers, 21 (75%) strongly agreed or agreed that the SRS was visible at appropriate times, and 18 (64%) strongly agreed or agreed that the SRS contains relevant information to assess surgical risk. 8 (29%) strongly agreed or agreed that the SRS was used to facilitate preoperative discussion. The 2023 survey additionally asked whether the SRS impacted the decision to operate; among providers who had interacted with the SRS, 4 (36%) strongly agreed or agreed with this statement. Conclusions: Integration of a real-time, MLM-generated SRS into the EHR and preoperative workflow can be successfully performed. While our integrated SRS is visible to surgeons and contains risk assessment felt relevant to the operation, the impact of the SRS on clinical care and decision making remains unclear.

DiaTracer enables spectrum-centric analysis of diaPASEF proteomics data.

Data-independent acquisition (DIA) has become a widely used strategy for peptide and protein quantification in mass spectrometry-based proteomics studies. The integration of ion mobility separation into DIA analysis, such as the diaPASEF technology available on Bruker's timsTOF platform, further improves the quantification accuracy and protein depth achievable using DIA. We introduce diaTracer, a new spectrum-centric computational tool optimized for diaPASEF data. diaTracer performs three-dimensional (m/z, retention time, ion mobility) peak tracing and feature detection to generate precursor-resolved "pseudo-MS/MS" spectra, facilitating direct ("spectral-library free") peptide identification and quantification from diaPASEF data. diaTracer is available as a stand-alone tool and is fully integrated into the widely used FragPipe computational platform. We demonstrate the performance of diaTracer and FragPipe using diaPASEF data from cerebrospinal fluid (CSF) and plasma samples, data from phosphoproteomics and HLA immunopeptidomics experiments, and low-input data from a spatial proteomics study. We also show that diaTracer enables unrestricted identification of post-translational modifications from diaPASEF data using open/mass offset searches.

Discovery of allosteric regulators with clinical potential to stabilize alpha-L-iduronidase in mucopolysaccharidosis type I.

Mucopolysaccharidosis type I (MPS I) is an inherited lysosomal disease caused by lowered activity of the enzyme alpha-L-iduronidase (IDUA). Current therapeutic options show limited efficacy and do not treat some important aspects of the disease. Therefore, it may be advantageous to identify strategies that could improve the efficacy of existing treatments. Pharmacological chaperones are small molecules that protect proteins from degradation, and their use in combination with enzyme replacement therapy (ERT) has been proposed as an alternative therapeutic strategy. Using the SEE-Tx® proprietary computational drug discovery platform, a new allosteric ligand binding cavity in IDUA was identified distal from the active site. Virtual high-throughput screening of approximately 5 million compounds using the SEE-Tx® docking platform identified a subset of small molecules that bound to the druggable cavity and functioned as novel allosteric chaperones of IDUA. Experimental validation by differential scanning fluorimetry showed an overall hit rate of 11.4%. Biophysical studies showed that one exemplary hit molecule GT-01803 bound to (Kd = 22 μM) and stabilized recombinant human IDUA (rhIDUA) in a dose-dependent manner. Co-administration of rhIDUA and GT-01803 increased IDUA activity in patient-derived fibroblasts. Preliminary in vivo studies have shown that GT-01803 improved the pharmacokinetic (PK) profile of rhIDUA, increasing plasma levels in a dose-dependent manner. Furthermore, GT-01803 also increased IDUA enzymatic activity in bone marrow tissue, which benefits least from standard ERT. Oral bioavailability of GT-01803 was found to be good (50%). Overall, the discovery and validation of a novel allosteric chaperone for rhIDUA presents a promising strategy to enhance the efficacy of existing treatments for MPS I. The compound's ability to increase rhIDUA activity in patient-derived fibroblasts and its good oral bioavailability underscore its potential as a potent adjunct to ERT, particularly for addressing aspects of the disease less responsive to standard treatment.

Application of an Automated Top Coal Caving Control System: The Case of Wangjialing Coal Mine

China has made notable advancements in the intelligent construction of coal mines. However, for longwall top coal caving (LTCC) mining faces, a key obstacle impeding the intelligent transition of the coal-cutting process is automated control. This paper focuses on the aforementioned issue and comprehensively considers the pre-, intra-, and post-coal-caving stages. In this work, diverse detection and monitoring technologies are integrated at various stages through a computer platform, facilitating the construction of an automated coal caving control system with self-perception, self-learning, self-decision-making, and self-execution capabilities. Key technologies include ground-penetrating radar-based top coal thickness detection, inertial navigation-based shearer positioning, tail beam vibration-based identification of coal and gangue, and magnetostrictive sensor-based monitoring of the tail beam and insert plate attitude. In this study, the 12309 working face of the Wangjialing Coal Mine was experimentally validated, and the efficacy of the aforementioned key technologies was assessed. The results demonstrated that the control requirements for automated coal caving are satisfied by the maximum errors. Automatic regulation of coal caving was realized through the implementation of this system, thereby facilitating initiation and cessation and yielding promising experimental outcomes. Overall, this system offers practical insights for intelligent construction in current LTCC mining faces and the sustainable development of coal resources.

Exploring the impact of biosecurity measures on somatic cell score in mountain dairy farms considering the CLASSYFARM system

Biosecurity plays a crucial role in preventing the introduction of infectious diseases to a herd as well as the spread of diseases within or between animals and herds. In particular, biosecurity measures are crucial for maintaining animal health and reducing the need for the application of antibiotic substances for fighting the rising antibiotic resistance. The object of this cross-sectional study was to investigate the presence of biosecurity measures and their association with milk quality parameters, with a special focus on somatic cell count (SCC) − an indicator for udder health −, in small-scale mountain dairy farms. Therefore, for the very first time, the CLASSYFARM system was considered, which is a computer platform integrated into the Italian national veterinary portal, that processes a significant amount of data from various sources collected in the field or from other information systems (e.g. animal welfare, health status, biosecurity, antimicrobial use, slaughterhouse information). A total of 169 dairy farms were included in the study. Biosecurity measures, based on 15 questions required in the CLASSYFARM welfare assessment protocol, as well as information about husbandry systems, milking systems and pasture practices were gathered and combined with milk yield data, provided by the South Tyrolean dairy association. Farms only scored 44.00 points on average in a scoring system from 0.00 to 100.00 points that was be able to summarize 15 different biosecurity measures in one index. Our results show a clear negative correlation (−0.713) between the biosecurity index and somatic cell score (SCS) indicating that a higher level of biosecurity, which reflects the presence of biosecurity measures within a farm, is associated with lower SCC levels. Furthermore, we found significant correlations between SCS and milk production (−0.629), confirming that udder health is linked to higher milk production. Fat, protein, and the fat-to-protein ratio showed a positive correlation with SCS (0.281, 0.146, 0.106), likely to be caused by a concentration shift effect (dilution effect). Husbandry system, breed, milking system, and pasture practices seem to have an impact as well, but the main factor was the biosecurity score. This study highlights the importance of implementing biosecurity measures for ensuring animal health and thus productivity and quality in milk production, even in small-scale farms, which are characterized by limited structure availability and smaller herds compared to big dairy enterprises in the lowlands.

Emerging Industrial Internet of Things Open-Source Platforms and Applications in Diverse Sectors

Revolutionary advances in technology have been seen in many industries, with the IIoT being a prime example. The IIoT creates a network of interconnected devices, allowing smooth communication and interoperability in industrial settings. This not only boosts efficiency, productivity, and safety but also provides transformative solutions for various sectors. This research looks into open-source IIoT and edge platforms that are applicable to a range of applications with the aim of finding and developing high-potential solutions. It highlights the effect of open-source IIoT and edge computing platforms on traditional IIoT applications, showing how these platforms make development and deployment processes easier. Popular open-source IIoT platforms include DeviceHive and Thingsboard, while EdgeX Foundry is a key platform for edge computing, allowing IIoT applications to be deployed closer to data sources, thus reducing latency and conserving bandwidth. This study seeks to identify potential future domains for the implementation of IIoT solutions using these open-source platforms. Additionally, each sector is evaluated based on various criteria, such as development requirement analyses, market demand projections, the examination of leading companies and emerging startups in each domain, and the application of the International Patent Classification (IPC) scheme for in-depth sector analysis.

Automated Python workflow for generating Sentinel-1 PSI and SBAS interferometric stacks using SNAP on Geospatial Computing Platform

Detecting and monitoring surface deformation using radar satellite data is vital in geohazard assessment. Sentinel-1 has provided unprecedented spatial and temporal resolution, but data processing is complicated and poses computational challenges. Although software and tools exist, each with its own limitations. SNAP-ESA is notable for its user-friendly interface and stable performance in Interferometric Synthetic Aperture Radar (InSAR). However, SNAP-ESA lacks a flexible approach for generating interferometric time series stacks for Persistent Scatterer Interferometry (PSI) and Small Baseline Subset (SBAS) techniques and faces computational challenges over large areas. Here, we present an automated Python workflow, SNAPWF, using SNAP-ESA to enable efficient PSI and SBAS interferometric time series stacks generation using flexible network graphs. SNAPWF has been implemented on a dedicated geospatial computing platform, enabling efficient performance over large areas. Results confirm its ability to generate PSI and SBAS interferometric stacks using full Sentinel-1 scenes and achieve results comparable to existing software.

Exploring memory synchronization and performance considerations for FPGA platform using the high-abstracted OpenCL framework: Benchmarks development and analysis.

A key benefit of the Open Computing Language (OpenCL) software framework is its capability to operate across diverse architectures. Field programmable gate arrays (FPGAs) are a high-speed computing architecture used for computation acceleration. This study investigates the impact of memory access time on overall performance in general FPGA computing environments through the creation of eight benchmarks within the OpenCL framework. The developed benchmarks capture a range of memory access behaviors, and they play a crucial role in assessing the performance of spinning and sleeping on FPGA-based architectures. The results obtained guide the formulation of new implementations and contribute to defining an abstraction of FPGAs. This abstraction is then utilized to create tailored implementations of primitives that are well-suited for this platform. While other research endeavors concentrate on creating benchmarks with the Compute Unified Device Architecture (CUDA) to scrutinize the memory systems across diverse GPU architectures and propose recommendations for future generations of GPU computation platforms, this study delves into the memory system analysis for the broader FPGA computing platform. It achieves this by employing the highly abstracted OpenCL framework, exploring various data workload characteristics, and experimentally delineating the appropriate implementation of primitives that can seamlessly integrate into a design tailored for the FPGA computing platform. Additionally, the results underscore the efficacy of employing a task-parallel model to mitigate the need for high-cost synchronization mechanisms in designs constructed on general FPGA computing platforms.

Organoids, Biocybersecurity, and Cyberbiosecurity—A Light Exploration

Organoids present immense promise for studying organ systems and their functionality. Recently, they have become the subject of exploration outside of purely biomedical uses in multiple directions. We will explore the rapidly evolving landscape of organoid research over the 21st century, discussing significant advancements in organoid research and highlighting breakthroughs, methodologies, and their transformative impact on our understanding of physiology and modeling. In addition, we will explore their potential use for biocomputing and harnessing organoid intelligence, investigate how these miniaturized organ-like structures promise to create novel computational models and processing platforms allowing for innovative approaches in drug discovery, personalized medicine, and disease prediction. Lastly, we will address the ethical dilemmas surrounding organoid research by dissecting the intricate ethical considerations related to the creation, use, and potential implications of these in vitro models. Through this work, the goal of this paper is to provide introductory perspectives and bridges that will connect organoids to cybersecurity applications and the imperative ethical discourse accompanying its advancements with commentary on future uses.

An insight into texture evolution and tailoring during multi-pass cold pilgering of high-strength titanium alloy tubular materials

The high-strength titanium alloy tubular materials are extensively applied in aviation and aerospace industries, wherein the mechanical properties, subsequent formability and in-service performance of the tube are strongly dependent on the texture variation. Cold pilgering has been a favorable fabrication technology for hard-to-deform tubes, however, the coupling effects of inherently low symmetric and complex multi-pass thermal-mechanical histories led to the significant diversity of the texture evolution of high-strength titanium alloy tubular materials. The traditional methods which mainly dependent on the Q value and deformation degree adjusting are difficult to achieve the precise tailoring of texture in multi-pass cold pilgering process. In this work, to break the limitation of the traditional method, by taking into account the mechanical properties iterating in each pass and the texture inheritance in recrystallization annealing process, a numerical computation platform for texture evolution prediction during multi-pass cold pilgering of titanium alloy tube was firstly established. The accuracy and reliability of the platform was verified by the 3-pass cold pilgering experiment and texture characterization. Secondly, the crucially effect of the initial texture was deeply explored in this work, the results indicated that the desired radial texture would be enhanced only when the Q value and area reduction ratio (AR) exceed the corresponding threshold under the specific initial texture, in which the tensile twinning was the vital deformation mode for the grain rotation and texture evolution. Thirdly, the quantitative correlation between processing conditions and finished radial texture intensity fRD was established which comprehensively introduced the Q value, AR and initial texture intensity. In contrast to the accuracy of 59.4% based on the single factor Q value and 75.1% based on double factor of Q value and AR, the prediction accuracy of the quantitative correlation established in this work upgraded to approximate to 95%. Finally, taking the contractile strain ratio (CSR) as the index to reflect the macro anisotropy mechanical properties, the bonds that unite the processing conditions, fRD and CSR were also built.

PROSCA: an online platform for humanized scaffold mining facilitating rational protein engineering.

Protein scaffolds with small size, high stability and low immunogenicity show important applications in the field of protein engineering and design. However, no relevant computational platform has been reported yet to mining such scaffolds with the desired properties from massive protein structures in human body. Here, we developed PROSCA, a structure-based online platform dedicated to explore the space of the entire human proteome, and to discovery new privileged protein scaffolds with potential engineering value that have never been noticed. PROSCA accepts structure of protein as an input, which can be subsequently aligned with a certain class of protein structures (e.g. the human proteome either from experientially resolved or AlphaFold2 predicted structures, and the human proteins belonging to specific families or domains), and outputs humanized protein scaffolds which are structurally similar with the input protein as well as other related important information such as families, sequences, structures and expression level in human tissues. Through PROSCA, the user can also get excellent experience in visualizations of protein structures and expression overviews, and download the figures and tables of results which can be customized according to the user's needs. Along with the advanced protein engineering and selection technologies, PROSCA will facilitate the rational design of new functional proteins with privileged scaffolds. PROSCA is freely available at https://idrblab.org/prosca/.

Graphene–fullerene heterostructures as robust and flexible nanomechanical bits

Electrical computers have revolutionized society over the past several decades, but questions have remained about their ability to perform in extreme environments, such as their stability at high-temperature conditions. This has motivated the recent surge of interest in developing mechanical computing platforms at all length scales, including the nanoscale, in which traditional electrical computers are augmented with mechanical ones. However, the most proposed nanomechanical bits are volatile memory bits based on the dynamic response of nanomechanical resonators, and as such, there is a need to develop robust and reprogrammable entirely non-volatile nanomechanical bits. Here, we exploit the multiple quasi-stable configurations of the graphene/fullerene/graphene (GFG) van der Waals heterostructure to work as a novel nonvolatile nanomechanical bit. The GFG heterostructure is unique and robust in that it can return to its initial state without further mechanical input, it can be used for logic functions at relevant operating temperatures through simple application of uniaxial strain, it can exhibit reprogrammability between five basic logic gates (NOT, AND, NAND, OR, NOR) by varying the magnitude or direction of applied strain, and it can be used to represent combinatorial logic through full and half adders. These findings provide a new opportunity to develop mechanical computers based on the large class of nanomaterials.

Real-time and Intelligent Moving Targets Tracking based on UAV Remote Sensing Video Camera and Brain-like Computing Chips

Abstract. Moving target tracking technology based on Unmanned Aerial Vehicles (UAV) is widely used in many fields such as automatic inspection and emergency response. The existing moving target tracking methods usually have the problems of large computation and low tracking efficiency. Limited by the computing power of the UAV platform, real-time tracking and analysis of multiple targets based on the video data collected by UAV platform is a difficult task. In this paper, we proposed a novel Target Specific Filtering Tracking with Memory (TSFMTrack) method designed for UAV-based real-time tracking tasks, which involves a Tracklet Filtering Module (TFM) for capturing object appearance features and a Tracklet Matching Module (TMM) for bounding box association in each frame. By experimental comparison with other State-Of-The-Art (SOTA) methods on popular MOT and UAV tracking datasets, the TSFMTrack have shown obvious advantages in accuracy, computational efficiency and reliability. Furthermore, we deployed the TSFMTrack on the brain-inspired chip Lynchip KA200, the experimental results have shown that the TSFMTrack is effective on edge computational platform and suitable for UAV real-time tracking tasks.

Highly 28Si enriched silicon by localised focused ion beam implantation

Solid-state spin qubits within silicon crystals at mK temperatures show great promise in the realisation of a fully scalable quantum computation platform. Qubit coherence times are limited in natural silicon owing to coupling to the 29Si isotope which has a non-zero nuclear spin. This work presents a method for the depletion of 29Si in localised volumes of natural silicon wafers by irradiation using a 45 keV 28Si focused ion beam with fluences above 1 × 1019 ions cm−2. Nanoscale secondary ion mass spectrometry analysis of the irradiated volumes shows residual 29Si concentration down to 2.3 ± 0.7 ppm and with residual C and O comparable to the background concentration in the unimplanted wafer. After annealing, transmission electron microscopy lattice images confirm the solid phase epitaxial re-crystallization of the as-implanted amorphous enriched volume extending over 200 nm in depth.

Predictions of Immunodominant Epitope Peptides From the AsaA Type VI Secretion System in Acinetobacter baumannii: A Computational Approach.

Introduction Acinetobacter baumannii, designated as a priority pathogen by the World Health Organization (WHO), is responsible for recalcitrant infections in immunocompromised patients. The type VI secretion system (T6SS) is a class of macromolecular secretion machines, contributing to its virulence. The aim of this study is thus to predict the immune-dominant epitope peptides from theAcinetobacter T6SS-associated protein of A. baumannii(AsaA). Methods AsaA protein retrieval from the bacteria was carried out using computational platforms and the evaluation of antigenicity and allergenicitywas performed. The T-cell epitopes of major histocompatibility complex class II binders were identified followed by molecular docking of the immune-dominant epitopes with human leukocyte antigen alleles using the ClusterPro server (https://cluspro.org/help.php). Additionally, the B-cell epitopes were predicted. Results Immune-informatic analysis showed immune-dominant peptides in the most favored regions with promising interactions with HLAalleles DP, DQ, DR, and toll-like receptor showing high binding capacity. Conclusion In the present investigation, epitope 1 (LILFLIGNY) was found to be a promising candidate for the synthesis of vaccines. However, it requires further experimentation for its immunological memory and response.

Development of an Integrated Computing Platform for Measuring, Predicting, and Analyzing the Profile-Specific Fixity of Railway Tracks

The current measures for railway track fixity in the UK’s railway remain at a relatively low level of granularity. This paper presents a pilot proof-of-concept study on the development of an integrated computing framework for improving the measurement, prediction, and analysis of profile-specific track fixity in the context of the UK’s rail network. The framework is aimed to produce a data integration and mining tool, which can determine track fixity parameters for any given section of track. In this study, we propose to measure track movement based on point cloud data and assess the track fixity by a set of parameters, such as the direction and rate of the track movement relative to the plane of the rail within a certain period. We seek to integrate a data mining algorithm into the framework to predict these parameters, given vast amounts of disparate and heterogeneous data of potential influencing factors in the area. From the study, we have developed a prototype framework, which allows the rapid implementation of data workflows with the necessary functionality. The feasibility of the prototype was demonstrated by training a random forest model on real data from an approximately 80-km section of the East Coast Main Line, southeast of Edinburgh, Scotland. The modeling results indicate that the curvature, cant, and maximum speed of trains are among the key factors that impact on, and are critical for predicting and analyzing, the profile-specific track fixity.

Regional-scale fault-to-structure earthquake simulations with the EQSIM framework: Workflow maturation and computational performance on GPU-accelerated exascale platforms

Continuous advancements in scientific and engineering understanding of earthquake phenomena, combined with the associated development of representative physics-based models, is providing a foundation for high-performance, fault-to-structure earthquake simulations. However, regional-scale applications of high-performance models have been challenged by the computational requirements at the resolutions required for engineering risk assessments. The EarthQuake SIMulation (EQSIM) framework, a software application development under the US Department of Energy (DOE) Exascale Computing Project, is focused on overcoming the existing computational barriers and enabling routine regional-scale simulations at resolutions relevant to a breadth of engineered systems. This multidisciplinary software development—drawing upon expertise in geophysics, engineering, applied math and computer science—is preparing the advanced computational workflow necessary to fully exploit the DOE’s exaflop computer platforms coming online in the 2023 to 2024 timeframe. Achievement of the computational performance required for high-resolution regional models containing upward of hundreds of billions to trillions of model grid points requires numerical efficiency in every phase of a regional simulation. This includes run time start-up and regional model generation, effective distribution of the computational workload across thousands of computer nodes, efficient coupling of regional geophysics and local engineering models, and application-tailored highly efficient transfer, storage, and interrogation of very large volumes of simulation data. This article summarizes the most recent advancements and refinements incorporated in the workflow design for the EQSIM integrated fault-to-structure framework, which are based on extensive numerical testing across multiple graphics processing unit (GPU)-accelerated platforms, and demonstrates the computational performance achieved on the world’s first exaflop computer platform through representative regional-scale earthquake simulations for the San Francisco Bay Area in California, USA.

Breaking the Barriers: Machine-Learning-Based c-RASAR Approach for Accurate Blood-Brain Barrier Permeability Prediction.

The intricate nature of the blood-brain barrier (BBB) poses a significant challenge in predicting drug permeability, which is crucial for assessing central nervous system (CNS) drug efficacy and safety. This research utilizes an innovative approach, the classification read-across structure-activity relationship (c-RASAR) framework, that leverages machine learning (ML) to enhance the accuracy of BBB permeability predictions. The c-RASAR framework seamlessly integrates principles from both read-across and QSAR methodologies, underscoring the need to consider similarity-related aspects during the development of the c-RASAR model. It is crucial to note that the primary goal of this research is not to introduce yet another model for predicting BBB permeability but rather to showcase the refinement in predicting the BBB permeability of organic compounds through the introduction of a c-RASAR approach. This groundbreaking methodology aims to elevate the accuracy of assessing neuropharmacological implications and streamline the process of drug development. In this study, an ML-based c-RASAR linear discriminant analysis (LDA) model was developed using a dataset of 7807 compounds, encompassing both BBB-permeable and -nonpermeable substances sourced from the B3DB database (freely accessible from https://github.com/theochem/B3DB), for predicting BBB permeability in lead discovery for CNS drugs. The model's predictive capability was then validated using three external sets: one containing 276,518 natural products (NPs) from the LOTUS database (accessible from https://lotus.naturalproducts.net/download) for data gap filling, another comprising 13,002 drug-like/drug compounds from the DrugBank database (available from https://go.drugbank.com/), and a third set of 56 FDA-approved drugs to assess the model's reliability. Further diversifying the predictive arsenal, various other ML-based c-RASAR models were also developed for comparison purposes. The proposed c-RASAR framework emerged as a powerful tool for predicting BBB permeability. This research not only advances the understanding of molecular determinants influencing CNS drug permeability but also provides a versatile computational platform for the rapid assessment of diverse compounds, facilitating informed decision-making in drug development and design.

Abstract PO2-07-04: Applying the Alliance Trial Guidelines in Multi-focal Breast Disease Using an Artificial Intelligence Computational Platform: Economic Analysis and Cosmetic Sensitivity

Abstract Background: Traditionally, multi-focal breast cancer results in mastectomy. The Alliance Trial offers a paradigm shift in surgical options available for multi-focal breast cancer patients in the context of adjuvant chemotherapy. In the trial, patients with multi-focal disease (< 3 tumors) who underwent breast conservation surgery (BCS) were found to have similar outcomes to patients undergoing mastectomy. BCS for large volume tumors ( >30%) has been cited as having a high potential for cosmetic defect, and hence represents a typical upper limit for potential tissue removal in BCS. Here, we evaluated a patient cohort to better understand the economic impact of the Alliance trial and further categorize patients that would most benefit without suffering cosmetic impact. We employed a novel computational technology to quantify the ratio of tumor size to breast tissue volume. Methods: Using a publicly available, single site cohort (n=243, DUMC) of breast cancer patients that underwent mastectomy, we segmented the tumors using our TumorSight Viz software platform. This platform uses artificial intelligence to segment the tumor and surrounding tissues and allows for a volumetric and morphologic assessment in 3D space. We then applied relevant inclusion/exclusion criteria from the Alliance Trial to the cohort (Saha et al, 2018). In trial eligible patients, we used TumorSight Viz to create a convex hull (CH) around the multi-focal disease using dilations of 1 cm and 2 cm. The volume of the CH, corresponding to proposed surgical extirpation, and the overall breast volume (BV) were then computationally assessed in 3D. The ratio of CH to BV (CH:BV) was calculated and a cutoff of 30% (high potential for cosmetic deformity) was applied. A cost analysis was then carried out. We determined the aggregate per annum savings that could potentially be realized by transforming a subset of mastectomies to BCS by tabulating total costs of mastectomy+reconstruction vs. BCS+WBI (whole-breast irradiation), as well as adjusted for relative rates of adjuvant therapy (~80%) across the nationwide patient population. Results: We found that 19.3% of adjuvant mastectomy patients were eligible for BCS based on Alliance Trial criteria. Of those, 68% had tumor CH:BV < 30% when using a 1 cm dilation around the tumor. When using a 2 cm dilation, 56% had tumor CH:BV < 30%. Together, these results indicate that of all adjuvant mastectomy patients, an estimated 10.8-13.1% are eligible for BCS based on volumetric measures of cosmetically acceptable breast tissue removal. Our economic analysis of BCS vs. mastectomy revealed an estimated $28,500 cost savings for patients with private insurance, suggesting that both decreased costs and improved quality-of-life (QOL) can be mutually aligned. By assessing the nationwide number of patients receiving adjuvant therapy for breast cancer, alongside the percentage potentially eligible for BCS using the above cosmetic defect analysis, we estimate that BCS conversion from mastectomy offers to provide a net savings of $300-350 million annually. Conclusion: The Alliance Trial guidelines unveiled the potential option of BCS in ~20% of patients with multi-focal disease in our cohort, demonstrating considerable cost-savings. Computational tools can further differentiate individuals who may not be best candidates for BCS in this setting, ensuring high QOL and informed decision-making. Citation Format: John Pfeiffer, Matthew Biancalana, Dorys Lopez-Ramos, Bradley Feiger, Anuja Antony. Applying the Alliance Trial Guidelines in Multi-focal Breast Disease Using an Artificial Intelligence Computational Platform: Economic Analysis and Cosmetic Sensitivity [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-07-04.

Abstract PS03-08: Spatial biomechanics determines fate in breast cancer

Abstract Background: With the recent advancements of spatial omics, it has become increasingly recognized that spatial analysis the of tumor architecture is key to decipher the heterogenous intratumoral relationships between different tumor components and provide better understanding of cancer therapy resistance, as well as help to identify potential targets for personalized therapy. The challenge of the discovery and implementation of such biomarkers still lays in the technological and methodological difficulties, and their translation into clinical practice. In parallel a major advancement in cancer research has been the recognition of the importance of cancer biomechanical properties in cancer progression, metastasis, and therapy response. Methods: We developed a new computational platform to identify biomechanical drivers of cancer outcome from spatial omics data. We used it to identify recurring spatial patterns of these markers within the tumor microenvironment, and define the spatial scale of heterogeneity of such patterns. Our dataset consists in 700 primary breast cancer baseline samples analyzed with two 30-marker imaging mass cytometry panels to identify cancer, immune, and stromal cells and structures and their biomechanical states. Our patient cohort includes 20+ years of follow up, with up to 6 samples per patient from 64 patients alive 12 years post diagnosis, and 49 patients lost to breast cancer deaths. Patients were treated with surgery, radiation, chemotherapy, endocrine therapy, or combinations of these post-surgery. Based on this approach, we derived a new three-parameter quantitative metric of preferential spatial co-localization between different cells or structures, and we use this metric to stratify patients into responders and non-responders with the aid of single and multivariate survival analysis. Finally, we include a correlation with Atomic Force Microscopy to identify the mechanical signature of these driving patterns in breast cancer. Results: Our work enabled the identification of tumor-immune-stromal spatial patterns that drive breast cancer outcome and their spatial scale. Among our results, we were able to further define spatial regions of hypoxia-driven EMT. We found these regions to be usually on the scale of 50 mm radius, and enriched in presence of unspecified immune cells. They are also more frequent in ER+ areas, with high NaKATPase activity. Our analysis also shed light on the controversy on the role of fibroblasts. We have found that the presence of a strong, spatially structured stroma, in fibroblast-rich regions is a strong predictor of positive outcome. Similarly, in our analysis collagen cross-linking emerges as a positive control factor in Vimentin-rich microenvironments, offering new interpretation to the role of the ECM structure. Importantly, our first principle approach allows for the identification of driving structures beyond heterogeneity, and thus enables easy extension to additional datasets. Finally, when correlating these patterns with Atomic force microscopy measurements, we can clearly see that patterns predictive of poor survival present a clear heterogeneous stiffness signature, as opposed to a very homogenous good survival pattern signature. Conclusion: These results confirm the importance of spatial distribution of biomechanical drivers in cancer, offering new avenues of physics-based therapeutic targets. Our results also offer a solid base to inform machine learning algorithms on how to significantly parametrize spatial patterns in breast cancer for clinical significance. Furthermore, we demonstrate for the first time the use of Atomic Force Microscopy as a single biomarker for these patterns in breast cancer with a direct correlation based on pathology matching. Citation Format: Sara Nizzero, Maria Pelaez Soni, Gregory Zaugg, Mariam Gachechiladze, Yitian Xu, Licheng Zhang, Junjun Zheng, Brian Menegaz, Lee B Jordan, Colin A Purdie, Philip R Quinlan, Chandandeep Nagi, Karla A Sepulveda, Philipp Oertle, Tobias A Appenzeller, Marko Loparic, Zhihui Wang, Shu-Hsia Chen, Vittorio Cristini, Marija Plodinec, Alastair Thompson. Spatial biomechanics determines fate in breast cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS03-08.