Internal Data Research Articles

Full-Section Deformation Monitoring of High-Altitude Fault Tunnels Based on Three-Dimensional Laser Scanning Technology

In traditional tunnel monitoring, the characteristic points of an object within a tunnel are measured to obtain information about the object. Considering the limitations of the traditional method in measuring the complex surface structure of tunnels, such as limited monitoring points, a long measurement period, and low precision, this study introduces an approach that uses three-dimensional (3D) laser scanning for monitoring tunnel cross-section deformation. Using this approach, the soft surrounding rock of a high-altitude ultralong tunnel was taken as the monitoring object. The test tunnel was first scanned using a 3D laser scanner, and the collected data were processed. The internal structural data of the tunnel were subsequently compared with its actual contour lines and the data of its primary branch and secondary lining on different dates. The results indicate that the arch roof of the tunnel tended to be stable within a certain time range when the positions of the primary branch and secondary lining were at different measuring points with different pile numbers. The deformation of the pile number on the left and right sides did not generally exceed 0.02 m, except at a few measuring points. A comparison between the actual cross section of the initial branch and that of the designed section showed that the actual elevation of the arch of the initial branch of the tunnel was greater than its designed elevation by no more than 0.3 m. Hence, through this study, a convenient and practical method is presented for monitoring deformation in complex curved tunnel structures.

Optimization of Internal Control for Budget Operations in Public Institutions Based on Random Forest Algorithm

In order to promote the construction of internal control in public institutions and improve work efficiency, the author proposes an optimization of internal control in public institution budget business based on random forest algorithm. We have constructed a big data audit framework for internal control of A Maritime Bureau based on the financial cloud platform and sorted out its audit process. By using the random forest algorithm to identify suspicious points in the internal control audit of administrative institutions at the data level, an example analysis is conducted using some data from A Maritime Bureau's assets, budget, revenue and expenditure, infrastructure, and contract business. The results indicate that the design of the internal control big data audit plan for administrative institutions will promote the innovation of audit information technology and application in A Maritime Bureau, provide theoretical guidance for the internal control big data audit carried out by administrative institutions, and effectively solve the problems of high workload and low work efficiency when A Maritime Bureau conducts internal control big data audits, thereby improving audit efficiency.

A Bayesian platform trial design with hybrid control based on multisource exchangeability modelling.

Enrolling patients to the standard of care (SOC) arm in randomized clinical trials, especially for rare diseases, can be very challenging due to the lack of resources, restricted patient population availability, and ethical considerations. As the therapeutic effect for the SOC is often well documented in historical trials, we propose a Bayesian platform trial design with hybrid control based on the multisource exchangeability modelling (MEM) framework to harness historical control data. The MEM approach provides a computationally efficient method to formally evaluate the exchangeability of study outcomes between different data sources and allows us to make better informed data borrowing decisions based on the exchangeability between historical and concurrent data. We conduct extensive simulation studies to evaluate the proposed hybrid design. We demonstrate the proposed design leads to significant sample size reduction for the internal control arm and borrows more information compared to competing Bayesian approaches when historical and internal data are compatible.

Optimizin Restaurant Tax Supervision In Jakarta: A Case Study Of The Tanah Abang Area

This research aims to explain the business process of supervising the restaurant tax collection in DKI Jakarta. This research also aims to provide suggestions to optimize supervising the restaurant tax collection in DKI Jakarta. The approach used in this research is a qualitative approach with descriptive methods. The data collection in this study was done using the techniques of observation, interview, and documentation study. The results of this study are an explanation of the general description of the restaurant tax collection supervision business process such as monitoring the payment of periodic restaurant tax deposits, comparison of third-party CMS data with internal restaurant taxpayer data, comparison of restaurant tax periodic deposit payment data with similar taxpayers (benchmarking), and implementation of silent operations to taxpayers business premises. This research also produced a proposal in the form of a method to optimize restaurant tax collection supervision activities in DKI Jakarta, especially in Tanah Abang area, namely by using tax data integration between central and local taxes, which is the result of cooperation between DKI Jakarta Provincial Government and the Directorate General of Taxes of the Ministry of Finance of the Republic of Indonesia. This research has several contributions, namely this research supplements the literature regarding the supervision of local tax collection, especially restaurant tax in Indonesia. In addition, the results of this study can also be used by DKI Jakarta Regional Revenue Agency (Bapenda DKI Jakarta) in the formulation of policies on the supervision of local tax collection, especially restaurant tax.

The role of analytical performance specifications in international guidelines and standards dealing with metrological traceability in laboratory medicine.

The goal of metrological traceability is to have equivalent results for a measurand in clinical samples (CSs) irrespective of the in-vitro diagnostic medical device (IVD-MD) used for measurements. The International Standards Organization standard 17511 defines requirements for establishing metrological traceability of values assigned to calibrators, trueness control materials and human samples used with IVD-MDs. Each step in metrological traceability has an uncertainty associated with the value assigned to a material. The uncertainty at each step adds to the uncertainty from preceding steps such that the combined uncertainty gets larger at each step. The combined uncertainty fora CS result must fulfil an analytical performance specification (APS) for the maximum allowable uncertainty (umax CS). The umax CS can be partitioned among the steps in a metrological traceability calibration hierarachy to derive the APS for maximum allowable uncertainty at each step. Similarly, the criterion for maximum acceptable noncommutability bias can be derived from the umax CS. One of the challenges in determining if umax CS is fulfilled is determining the repeatability uncertainty (u Rw) from operating an IVD-MD within a clinical laboratory. Most of the current recommendations for estimating u Rw from internal quality control data do not use a sufficiently representative time interval to capture all relevant sources of variability in measurement results. Consequently, underestimation of u Rw is common and may compromise assessment of how well current IVD-MDs and their supporting calibration hierarchies meet the needs of clinical care providers.

Impacts of unemployment benefit program on job search duration: Evidence from Indonesia

Purpose ― This study examines the impacts of the Unemployment Benefit Program (Jaminan Kehilangan Pekerjaan, JKP) on the duration of job search in Indonesia using internal data sourced from the Employment Social Security Administering Agency (Badan Penyelenggara Jaminan Sosial Tenaga Kerja, BPJS TK).Method ― The study employs the Regression Kinked Design (RKD) model and Ordinary Least Square (OLS) estimation method to analyze the data on laid-off workers who have received the JKP benefits.Findings ― The study finds that, on average, the JKP beneficiaries have a longer duration of job search than laid-off workers who do not receive benefits. An increase in the replacement rate of the JKP benefits is associated with an increased duration of both job search and benefit claims. The RKD plot indicates that a replacement rate lower than 45% may decrease the duration of benefit claims and job searches. Implication ― The study recommends that the government should consider two potential options: (1) maintaining the current level of monetary benefits that leads to a longer job search duration; in other words, the program management must continue providing benefits for the maximum duration offered (up to 6 months), or, (2) reducing the monetary benefits, if BPJS TK aims to reduce the job duration. The results may change if a more robust labor market information system is available.Originality ― This study represents the first attempt to investigate the impact of the Unemployment Benefit Program (JKP) on job search duration in Indonesia. It might also be important to enrich the existing literature on the impact of similar programs in developing countries and countries with large populations.

Refining outcome prediction after traumatic brain injury with machine learning algorithms.

Outcome after traumatic brain injury (TBI) is typically assessed using the Glasgow outcome scale extended (GOSE) with levels from 1 (death) to 8 (upper good recovery). Outcome prediction has classically been dichotomized into either dead/alive or favorable/unfavorable outcome. Binary outcome prediction models limit the possibility of detecting subtle yet significant improvements. We set out to explore different machine learning methods with the purpose of mapping their predictions to the full 8 grade scale GOSE following TBI. The models were set up using the variables: age, GCS-motor score, pupillary reaction, and Marshall CT score. For model setup and internal validation, a total of 866 patients could be included. For external validation, a cohort of 369 patients were included from Leuven, Belgium, and a cohort of 573 patients from the US multi-center ProTECT III study. Our findings indicate that proportional odds logistic regression (POLR), random forest regression, and a neural network model achieved accuracy values of 0.3-0.35 when applied to internal data, compared to the random baseline which is 0.125 for eight categories. The models demonstrated satisfactory performance during external validation in the data from Leuven, however, their performance were not satisfactory when applied to the ProTECT III dataset.

Geospatial Science for the Environmental Epidemiology of Cancer in the Exposome Era.

Geospatial science is the science of location or place that harnesses geospatial tools, such as geographic information systems (GIS), to understand the features of the environment according to their locations. Geospatial science has been transformative for cancer epidemiologic studies through enabling large-scale environmental exposure assessments. As the research paradigm for the exposome, or the totality of environmental exposures across the life course, continues to evolve, geospatial science will serve a critical role in determining optimal practices for how to measure the environment as part of the external exposome. The objectives of this article are to provide a summary of key concepts, present a conceptual framework that illustrates how geospatial science is applied to environmental epidemiology in practice and through the lens of the exposome, and discuss the following opportunities for advancing geospatial science in cancer epidemiologic research: enhancing spatial and temporal resolutions and extents for geospatial data; geospatial methodologies to measure climate change factors; approaches facilitating the use of patient addresses in epidemiologic studies; combining internal exposome data and geospatial exposure models of the external exposome to provide insights into biological pathways for environment-disease relationships; and incorporation of geospatial data into personalized cancer screening policies and clinical decision making.

Laboratory modelling of nonlinear power take-off damping and its effects on an offshore stationary cylindrical OWC device

The accurate evaluation of the wave energy conversion of oscillating water column (OWC) wave energy converter has consistently plagued scholars. In this paper, an original and well-validated test platform is proposed for exploring the flow characteristic of a nonlinear power take-off (PTO) damping, with which a more accurate and stable measurement of captured energy is obtained in the laboratory modelling. This study utilizes ten orifices to comprehensively investigate the impact of the PTO damping on an offshore stationary cylindrical OWC device. The joint effect of the PTO damping with wave height (Hi) or draft of OWC (Dr) is examined. The damping coefficient Cf = 10350 is found to be the optimal damping coefficient, in terms of both the peak wave energy conversion and the efficient bandwidth. It is revealed that the variations of the PTO damping can regulate the effect of Dr and Hi on wave energy conversion. An appropriate reduction in PTO damping can alleviate the weak energy output caused by the increased Dr and Hi. The effect of increased Hi on wave energy conversion could be judged by the size relation of the maximum and minimum values of internal pressure data.

113 Creation of an undergraduate certificate program in clinical and translational science following a six-step curriculum development process

OBJECTIVES/GOALS: Academic research centers often struggle to recruit and retain a diverse and competent clinical and translational science (CTS) workforce. Specifically, the clinical research professional (CRP) career pathway is not well known to undergraduate students and other individuals outside of academic medicine despite various potential career routes. METHODS/STUDY POPULATION: To address these workforce challenges, the CRP Task Force at the University of Cincinnati (UC) aims to train a competent and diverse CRP workforce through targeted educational programming in the UC undergraduate population. Using a six-step curriculum development process that included: 1) performing a needs assessment, 2) determining content, 3) writing goals and objectives, 4) selecting the educational strategies, 5) implementing the curriculum, and 6) evaluating the curriculum, we designed an undergraduate certificate program in CTS. RESULTS/ANTICIPATED RESULTS: The needs assessment included both internal and external data gathering to inform curriculum development and program decisions. Content was determined using the CRP Competency Framework 2.0, and program learning outcomes were written with both the competency framework and local workforce needs in mind. Educational strategies were selected based upon optimization of available resources and local expertise with an emphasis on interactive didactics complemented by experiential learning. Implementation is underway and evaluation will follow once students begin enrolling. DISCUSSION/SIGNIFICANCE: We anticipate an increase in numbers of well-qualified, diverse applicants who pursue CRP careers locally and regionally. In addition, we expect that the certificate program will build competency earlier in CRP staff, improving job satisfaction and retention as a result of a stronger foundation from which to build their professional skills.

Improving Customer Data Quality on Pension Bank Loans in Palembang-South Sumatra

The bank has transferred the retirement loan application service to the cooperative as one of its subsidiaries. The phenomenon that often occurs is misinformation from cooperative officers when explaining the number of credit applications that should be made. The purpose of this research is to find the right pattern and form for applying the latest data technology so that it can provide the right information in the credit application process. Measurement and analysis of research data are qualitative, with a descriptive explanation level. Viewed by place and classified as field research. Types and research strategies, including Case Studies. Retrieval, collection, and analysis techniques were carried out repeatedly and simultaneously until the data was saturated and credible. The research yields findings about internal customer data connection patterns that are accurate and up-to-date. The form of credit data management by providing recommendations and internal references to customer performance, so that process and procedure connectivity is guaranteed between Bank Financial Institutions (LKB) and Non-Bank Financial Institutions (LKBB). The conclusion from the research is that the track records or data records of retired customers at the bank must be properly and up-to-date so that when applying for the same credit at a subsidiary of the institution, there is no misunderstanding of information and guaranteed correct and correct data.

EXPRESS: Amplifying Off-Site Purchases with On-Site Retail Media Advertising

Retail media is of growing interest to brands, and online retailers offer a large collection of product information that is freely available 24/7 for consumers' online and offline purchase decisions. In this study, the authors map webrooming across ten product categories by combining internal data from consumer browsing behavior at an online retailer and survey responses from the same customers about their offsite behavior. Moreover, they build a random forest machine-learning model to predict the magnitude of webrooming across categories. They find that webrooming is economically substantial. On average, for every 10 customers who research the product category on-site and buy on-site, 17 other customers research the product category on-site and buy off-site. As to retail media's impact, upper-funnel and always-on online ad forms are associated with a higher number of online searchers in the laptop category that buy the brand offline. Beyond finding support for their propositions, the authors provide directions for future research on the cross-channel effects of retail media and how they generalize across brands, categories and retailers.

Short-term cooling load prediction for central air conditioning systems with small sample based on permutation entropy and temporal convolutional network

In air-conditioning load predicting, the mainstream methods mainly include mechanism model based methods and data-driven prediction methods. The mechanism model can achieve accurate prediction, but are normally used to predict non-existent buildings and by default have to make several assumptions regarding post occupancy. This study proposes a novel data-driven short-term cooling load prediction method based on historical data of the existing buildings to improve the prediction performance when the size of the training sample datasets, including load data and air conditioning internal operation data, is small. Firstly, an improved version of variational mode decomposition (IVMD) technique is proposed to deconstruct complex historical cooling load signals. Then, the permutation entropy and Savitzky-Golay (PE-SG) smoothing method is developed to extract singles with distinct characteristics and high regularity, which comprises multiple single-component amplitude and frequency modulation (AFM) signals, thus mitigating the interference of noise in predictive accuracy. Subsequently, a data-driven model based on temporal convolutional network (TCN) is constructed, which effectively alleviates the risks of over-fitting and under-fitting. Finally, the efficacy of the proposed method is validated through two case studies, and the obtained results show that the issues confronted by small sample datasets are adeptly solved by this method. The precision of prediction is significantly enhanced according to the comparison to other traditional models, showcasing greater stability and generalization ability.

Pengurangan Impor Sodium Klorat di Indonesia melalui Desain ClO2 Semi-Integrated pada Modifikasi ClO2 R8 Plant 10 Ton/Hari

Pulp production always has a bleaching process that uses chlorine dioxide (ClO2) as a bleaching pulp agent. One of the ClO2 production units that are used in the Pulp Industry in Indonesia is a ClO2 R8 non-integrated plant with the process of reacting sodium chlorate, sulfuric acid, and methanol. This ClO2 production type has a weakness, especially in supplying the raw material sodium chlorate, which is needed to import, and the characteristics are easy to explode, making production costs high. This research aims to modify the ClO2 R8 non-integrated by changing to a ClO2 semi-integrated plant. In this research, HYSYS was used to simulation to get mass and energy balance and other component data, and the design of equipment based on the American Society of Mechanical Engineers (ASME), Kern, internal data company, and others to get the specification equipment and also drawn using Autocad. The last step is to account for the economic for getting cost of production before and after modification, fixed capital cost, benefit, payback period, and return on investment. The resulting area, the first, modification of the ClO2 R8 plant by changing to ClO2 semi-integrated could produce chlorate solution 510 g/l which they use electrolyzers system from Chemetics with 20 cell M25 and hypo system. Second, the cost production for the ClO2 semi-integrated plant is 1393 USD/T, the benefit is 599.79 USD/year, the Payback period (PBP) is 6,65 years with an investment of about 3988 KUSD and RoI 15%.

Process-Structure-Property Relationship Development in Large-Format Additive Manufacturing: Fiber Alignment and Ultimate Tensile Strength.

Parts made through additive manufacturing (AM) often exhibit mechanical anisotropy due to the time-based deposition of material and processing parameters. In polymer material extrusion (MEX), printed parts have weak points at layer interfaces, perpendicular to the direction of deposition. Poly(lactic acid) with chopped carbon fiber was printed on a large-format pellet printer at various extrusion rates with the same tool pathing to measure the fiber alignment with deposition via two methods and relate it to the ultimate tensile strength (UTS). Within a singular printed bead, an X-ray microscopy (XRM) scan was conducted to produce a reconstruction of the internal microstructure and 3D object data on the length and orientation of fibers. From the scan, discrete images were used in an image analysis technique to determine the fiber alignment to deposition without 3D object data on each fiber's size. Both the object method and the discrete image method showed a negative relationship between the extrusion rate and fiber alignment, with -34.64% and -53.43% alignment per extrusion multiplier, respectively, as the slopes of the linear regression. Tensile testing was conducted to determine the correlation between the fiber alignment and UTS. For all extrusion rates tested, as the extrusion multiplier increased, the percent difference in the UTS decreased, to a minimum of 8.12 ± 14.40%. The use of image analysis for the determination of the fiber alignment provides a possible method for relating the microstructure to the meso-property of AM parts, and the relationship between the microstructure and the properties establishes process-structure-property relationships for large-format AM.

Effects of Electronic Nicotine Delivery Systems Substitution on Body Weight Status: Protocol for a Systematic Review and Meta-Analysis.

Weight gain following smoking cessation is a well-documented concern, often attributed to the absence of nicotine's metabolic influence. The adoption of Electronic Nicotine Delivery Systems (ENDS) has been used to achieve smoking cessation, with claims of aiding weight control. However, existing reviews present conflicting conclusions on ENDS' impact on weight status, necessitating a rigorous evaluation. We aim to conduct a systematic review with meta-analysis to assess the actual impact of ENDS on weight status in individuals who have ceased or reduced conventional smoking. The primary goal is to provide clinicians with evidence-based insights into the potential effects of ENDS use as a smoking substitute on weight control. Adhering to PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols) guidelines, our systematic review will analyze randomized and nonrandomized controlled trials, clinical trials (quasi-experimental), and prospective or retrospective cohort studies on the weight status effects of ENDS among individuals who have either quit or reduced smoking. Searches will include PubMed, Scopus, and Cochrane Library, covering the period from 2010 to January 2024. A gray literature search and supplementary searches will be performed. Data will be extracted independently by 2 reviewers and quality assessments will be conducted concurrently. Quality assessments will use Joanna Briggs Institute tools, 2020 version, along with bias assessments for internal validity and reporting bias based on the Catalogue of Bias. The included studies will be examined for any internal data reporting discrepancies by using Puljak's checklist. Meta-analysis and subgroup analyses (ie, general ENDS usage, ENDS use coupled with a reduction in smoking exceeding 50%, and exclusive ENDS use for achieving smoking cessation) are planned. Certainty of evidence will be evaluated using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework. The protocol has been registered in PROSPERO (CRD42023494974) and the entire systematic review is expected to be completed by April 2024. The main goal of this review is to retrieve all current human research studies investigating the influence of ENDS on weight management among individuals who have quit or reduced smoking. Furthermore, the review will assess the quality of these studies and examine potential biases to identify the most dependable evidence available. Dissemination strategies will include traditional journal publications, social media announcements, and a white paper. The latter, available for download and distributed at conferences, aims to reach a broad audience, including clinicians and ENDS users. The review will address the importance of informing health care professionals and patients about the current and robust evidence regarding the effects of transitioning to ENDS for smoking cessation on weight status. PROSPERO CRD42023494974; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=494974. PRR1-10.2196/56324.

Abstract 1188: Identification of CXCL10-inducing chemotherapy/targeted therapy combinations for PD-1 blockade sensitization in “cold” triple negative breast cancer

Abstract Metastatic triple-negative breast cancer (mTNBC) currently has the poorest prognosis among subtypes, with limited treatment options due to molecular characteristics that render conventional targeted therapies ineffective. Immunotherapy, specifically anti-PD-(L)1 (ICIs), has revolutionized certain cancer treatments, but its efficacy in mTNBC has been disappointing. To enhance tumor responses to ICIs, combining them with immunogenic chemotherapy has been explored. Studies like IMpassion130 and KEYNOTE 522 indicate benefits of chemo-immunotherapy based on taxanes, especially in PD-L1+ tumors. PD-L1 expression often correlates with TILs infiltration, particularly cytotoxic T lymphocytes (CTLs), indicating an activated immune response in the tumor. However, tumors with poor CTL infiltration, known as "Cold" phenotype, are often associated with a lack of induction of CXCR3-associated chemokines (CXCL9, 10, and 11), crucial for CTLs recruitment in the tumor microenvironment. The limited efficacy of chemo-immunotherapy in PD-L1-negative tumors may stem from chemotherapy's inability to induce immunogenic cancer cell death, CXCR3-associated chemokines, and CTL recruitment. Internal data from a 4T1 TNBC murine model demonstrated that the taxane paclitaxel failed to induce CXCL10, CTL recruitment, and sensitization to PD-L1 blockade. The preclinical research project aimed to identify a combination therapy, inspired by current clinical practices, capable of inducing CXCL10 expression and restoring CTL recruitment in the 4T1 "cold" mTNBC model. We initially screened standard TNBC chemotherapies and selected carboplatin for its in vitro potential to induce CXCL10 production and its clinical use in association with ICIs in the TNBC. Combining carboplatin with a library of over 400 targeted therapies, we identified histone deacetylase (HDAC) inhibitors as significantly enhancing CXCL10 secretion. In vivo, this combination demonstrated therapeutic additivity, increasing immune recruitment, particularly CD8+ T lymphocytes, with the therapeutic effect reduced by CXCR3 blockade. Mechanistic exploration in vitro revealed that the combination activated the type I interferon pathway, inducing CXCL10 secretion. In vivo confirmation validated the role of the type I interferon pathway and CXCL10 in the combination's efficacy. This data outlines a chemotherapy and HDAC inhibitor combination capable of inducing CTL recruitment via increased CXCL10 production. Future project phases will focus on understanding HDACi's mechanism of action on CXCL10 production and the combination's potential to sensitize tumors to ICIs. Overall, this project underscores the role of epigenetic modifications in amplifying carboplatin-induced CXCL10 secretion, potentially converting "cold" ICI-resistant tumors into a "hot" phenotype. Citation Format: Laura Kalfeist, Stacy Petit, Loick Galland, Cyriane Poirrier, Romain Aucagne, François Ghiringhelli, Sylvain Ladoire, Emeric Limagne. Identification of CXCL10-inducing chemotherapy/targeted therapy combinations for PD-1 blockade sensitization in “cold” triple negative breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1188.

Abstract 908: SeNMo: A self-normalizing deep learning model for enhanced multi-omics data analysis in oncology

Abstract Multi-omics research has enhanced our understanding of cancer heterogeneity and progression. Investigating molecular data through multi-omics approaches is crucial for unraveling the complex biological mechanisms underlying cancer, thereby enabling more effective diagnosis, treatment, and prevention strategies. However, predicting patient outcomes through integration of all available multi-omics data is still an under-study research direction. Here, we present SeNMo (Self-normalizing Network for Multi-omics), a deep neural network that ensures the zero mean and unit variance of activations across network layers using the self-normalizing technique. Such normalizing techniques are critical in stable and robust learning of deep learning models. SeNMo is particularly efficient in handling multi-omics data characterized by high-width (many features) and low-length (fewer samples) attributes. We trained SeNMo for the task of overall survival of patients using pan-cancer multi-omics data involving 28 cancer sites from the Genomic Data Commons (GDC). The training multi-omics data includes gene expression, DNA methylation, miRNA expression, and protein expression modalities. We tested the model's performance on the Moffitt Cancer Center's internal data involving RNA expression and protein expression data. We evaluated the model’s performance in predicting patient’s overall survival using the concordance index (C-Index), which provides a robust measure of the model's predictive capability. SeNMo performed consistently well in the training regime, reflected by the validation C-Index≥0.6 on GDC's public data. In the testing regime on Moffitt's private data, SeNMo performed with a C-Index of 0.68. The model's performance increased when tested on low-dimensional data or when tested on single omic data such as RNA or protein expression data with a C-Index of 0.7. SeNMo proved to be a mini-foundation model for multi-omics oncology data because it demonstrated robust performance, adaptability across molecular data types, and universal approximator capabilities for the scale of molecular data it was trained on. SeNMo can be further scaled to any cancer site and molecular data type. It can also be fine-tuned for other downstream tasks such as treatment response prediction, risk stratification, patient subgroup identification, and others. Its ability to accurately predict patient outcomes and adapt to various downstream tasks indicates a new era in cancer research and treatment. For future research, SeNMo offers a powerful tool for uncovering deeper insights into the complex nature of cancer and sets a precedent for how artificial intelligence can be leveraged to handle the vast and intricate data in the biomedical field. We believe SeNMo and similar models are poised to transform the oncology landscape, offering hope for more effective, efficient, and patient-centric cancer care. Citation Format: Asim Waqas, Aakash Tripathi, Sabeen Ahmed, Ashwin Mukund, Paul Stewart, Mia Naeini, Hamza Farooq, Ghulam Rasool. SeNMo: A self-normalizing deep learning model for enhanced multi-omics data analysis in oncology [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 908.

Abstract 4924: Assessing microsatellite instability dominance in colorectal cancer phenotype: A multi-study initiative using multi-target transformers for genomic biomarker prediction

Abstract We used deep learning (DL) to predict multiple genetic biomarkers of colorectal cancer (CRC) from routine histopathology images, aiming to identify morphological changes associated with the mutational profile of tumors. A new Transformer-based prediction model was developed which can simultaneously predict multiple biomarkers from a single histopathology image. We compared the performance of the multi-target approach to conventional single-target models. We analyzed five patient cohorts from the Genetics and Epidemiology of CRC Consortium (GECCO) with colorectal cancer (N=1,385 patients). Our model was trained on an internal data set (N=739) to predict the presence of >200 genetic alterations, focusing on genes with non-silent mutations and mutational signatures, which were assessed using targeted sequencing. We used a 7-fold cross validation and validated the model on an independent, external data set (N=646). Morphological features of microsatellite instability (MSI) are detectable by DL models in histopathology images, making MSI a potential confounding factor. Thus, we also assessed whether our model can predict genetic alterations regardless of MSI status, or whether the model identifies the MSI phenotype. For the majority of biomarkers, the multi-target transformer reached a higher prediction performance as measured by the mean Area Under the Receiver Operating Characteristic (AUROC) Curve and lower standard deviation than single-target DL models. For the external validation set, the mean AUROC of the multi-target versus the single-target model was 0.78 (+/- 0.01) versus 0.72 (+/- 0.06) for BRAF mutations, 0.88 (+/- 0.01) versus 0.86 (+/- 0.03) for hypermutation status, 0.94 (+/- 0.01) versus 0.91 (+/- 0.02) for MSI, 0.86 (+/- 0.01) versus 0.80 (+/- 0.05) for RNF43 mutations and 0.72 (+/- 0.02) versus 0.69 (+/- 0.05) for TP53 mutations, respectively. However, our analysis of the individual prediction scores suggests that the model does not identify the phenotype of the alterations themselves. Instead, the model seems to determine the status of specific genetic alterations based on their correlation with MSI. Consequently, biomarkers which are associated with microsatellite stability have prediction scores negatively correlated with predicted MSI scores. Conversely, biomarkers which are associated with MSI have prediction scores positively correlated with predicted MSI scores. Our results demonstrate potential benefits of multi-target DL models in improving the predictive power and efficacy of histology-based biomarker extraction compared to single-target DL models. However, our results crucially highlight the importance of considering potential confounders, such as MSI status in CRC, and emphasize the need for extensive analysis beyond AUROC in studies focusing on DL-based biomarker detection. Citation Format: Marco Gustav, Marko van Treeck, Zunamys I. Carrero, Chiara M. Loeffler, Nic G. Reitsam, Bruno Märkl, Asier Rabasco Meneghetti, Lisa A. Boardman, Amy J. French, Ellen L. Goode, Andrea Gsur, Stefanie Brezina, Marc J. Gunter, Neil Murphy, Paul Limburg, Stephen Thibodeau, Sebastian Foersch, Robert Steinfelder, Tabitha Harrison, Ulrike Peters, Amanda Phipps, Jakob N. Kather. Assessing microsatellite instability dominance in colorectal cancer phenotype: A multi-study initiative using multi-target transformers for genomic biomarker prediction [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4924.

Patient-oriented research: An essential driver of learning health system capacity development.

Canada's health system faces a lag in implementing high-quality evidence and research-driven innovation into service delivery, while demonstrating accountability and benefit to the public. To address these challenges, Patient-Oriented Research (POR) builds teams that engage researchers, healthcare providers, decision-makers, and most importantly, patients (people with lived and living experience) in the process of generating and applying evidence to inform health services and decision-making. A Learning Health System (LHS) systematically integrates external evidence with internal data and experience and puts that knowledge into practice in a continuous cycle. Using a POR/LHS example from a BC health authority, we describe nine enablers required to support LHS capacity development. The LHS case study, Walk With Me, addresses a health system high-priority topic: the toxic drug crisis. Understanding the value of LHSs, along with the enablers required to support and implement them, will empower health leaders to champion and orchestrate positive change.