Blind Set Research Articles

Multi-site blinded validation of a deep learning approach for clinical-grade MSI/dMMR detection in colorectal cancer from H&E-stained pathology images.

44 Background: Testing for microsatellite instability (MSI) or mismatch repair deficiency (dMMR) is part of the diagnosis and clinical management of patients with colorectal cancer (CRC). Healthcare services recommend MSI or dMMR testing for all CRC patients to guide therapeutic choices and assist in identifying Lynch Syndrome. However, in clinical practice, high costs and the demand for timely test results, combined with the rising prevalence of CRC and a shrinking pathology workforce, present a barrier to universal adoption. This highlights the need for rapid and affordable alternatives. PANProfiler CRC (PPC) is a deep learning-based solution for detecting MSI/dMMR in CRC tumors that only requires whole slide images (WSIs) of haematoxylin and eosin (H&E)-stained tissue to provide test results. Using only WSIs, PPC offers an efficient alternative to standard testing. This study evaluates PPC's performance in a multi-site blinded setting. Methods: Blinded validation was performed using 3246 WSIs of H&E-stained CRC specimens. PPC provided outputs as "Stable", "Unstable", or "Indeterminate", with "Unstable" indicating dMMR or MSI-High, and "Stable" indicating proficient mismatch repair or non-MSI-High. "Indeterminate" was returned when PPC did not have a definitive result. PPC was evaluated by comparison to standard MSI/dMMR tests. Validation data spanned three cohorts from two sites (Table). St James’s University Hospital (SJUH), Leeds, UK, supplied Cohorts 1 and 2; Cohort 3 was sourced from Wales Cancer Biobank (WCB), UK. Blinded analysis was performed at SJUH. Results: Results are given (Table). PPC demonstrated an overall percent agreement of 93.91%, a positive percent agreement of 92.17%, and a negative percent agreement of 94.15%, returning a definitive result for 88.05% of WSIs. Conclusions: This real-world, multi-site, blinded validation study demonstrates PPC’s remarkable performance, comparable to standard tests for detecting MSI/dMMR in CRC, with high test replacement rates. In the clinical setting, PPC could significantly accelerate testing and enable timely delivery of stratified treatment plans. This accurate and cost-effective diagnostic solution promises to revolutionize MSI/dMMR testing in CRC. Blinded validation results of PPC with confidence intervals (CI) at 95%. Site Cohort Sample Size (Unstable; Stable) Overall Percent Agreement % (CI) Positive Percent Agreement % (CI) Negative Percent Agreement %(CI) Test Replacement Rate % SJUH 1 488 (78; 410) 92.79 (89.86-95.08) 90.16 (79.81-96.30) 93.24(90.11-95.62) 85.25 SJUH 2 2704 (318; 2386) 94.31(93.31-95.21) 92.31(88.48-95.18) 94.57(93.52-95.50) 88.42 WCB 3 54 (11; 43) 84.31 (71.41-92.98) 100.00(71.51-100.00) 80.00(64.35-90.95) 94.44 All 3246 (407; 2839) 93.91 (92.97-94.76) 92.17(88.82-94.78) 94.15(93.16-95.04) 88.05

Discrimination of Benign and Malignant Thyroid Nodules through Comparative Analyses of Human Saliva Samples via Metabolomics and Deep-Learning-Guided Label-free SERS.

Thyroid nodules are a very common entity. The overall prevalence in the populace is estimated to be around 65-68%, among which a small portion (less than 5%) is malignant (cancerous). Therefore, it is important to discriminate benign thyroid nodules from malignant thyroid nodules. In this study, an equal number of participants with benign and malignant thyroid nodules (N = 10/group) were recruited. Saliva samples were collected from each participant, and SERS spectra were acquired, followed by validation using a metabolomics approach. An additional equal number of patients (N = 40/group) were recruited to construct diagnostic models. The performance of various machine learning (ML) algorithms was assessed using multiple evaluation metrics. Finally, the reliability of the optimal model was tested using blind test data (N = 10/group for benign and malignant thyroid nodules). The results showed a consistent trend between the SERS metabolic profile and the metabolites identified through MS analysis. The Multi-ResNet algorithm was optimal, achieving a 95% accuracy in sample discrimination. Additionally, blind test data sets yielded an overall accuracy of 83%. In summary, the deep-learning-guided SERS technique holds great potential in the accurate discrimination of benign and malignant thyroid nodules via human saliva samples, which facilitates the noninvasive diagnosis of malignant thyroid nodules in clinical settings.

Optimizing Bone Regeneration with Demineralized Dentin-Derived Graft Material: Impact of Demineralization Duration in a Rabbit Calvaria Model.

This study evaluated the regenerative potential of demineralized dentin-derived matrix (DDM) as a bone graft material in rabbit calvaria. DDM, sourced from extracted teeth, is emerging as an alternative to traditional grafts like allografts and xenografts. We aimed to identify the most effective demineralization protocol to optimize the regenerative capacity of DDM. Four groups were compared: a control group without grafts, a non-demineralized DDM group, and two demineralized DDM groups (15 and 30 min demineralization). Histomorphometric analysis was conducted in a randomized and blinded setting at 2, 4, and 8 weeks post-graft placement. The results revealed that the 15 min demineralized DDM group showed the most significant new bone formation (42.51% ± 6.40% at 8 weeks; p < 0.05), suggesting its potential as a highly effective regenerative graft material.

Using machine learning approaches to develop a fast and easy-to-perform diagnostic tool for patients with light chain amyloidosis: a retrospective real-world study.

Immunoglobulin light chain (AL) amyloidosis is a severe disorder caused by the accumulation of amyloid fibrils, leading to organ failure. Early diagnosis is crucial to prevent irreversible damage, yet it remains a challenge due to nonspecific symptoms that often appear later in the disease progression. A retrospective study analyzed data collected from 133 AL amyloidosis patients and 271 non-AL patients with similar symptoms but different diagnoses between January 1st, 2017, and September 30th, 2022. Demographic data and laboratory test results were collected. Subsequently, significant features were identified by both logistic regression and independent expert clinical ability. Eventually, logistic regression and four machine learning (ML) algorithms were employed to construct a diagnostic model, utilizing fivefold cross-validation and blind set testing to identify the optimal model. The study successfully identified nine independent predictors of AL amyloidosis patients with kidney or cardiac involvement, respectively. Two models were developed to identify key features that distinguish AL amyloidosis from nephrotic syndrome and hypertrophic cardiomyopathy, respectively. The light gradient boosting machine (LightGBM) model emerged as the most effective, demonstrating superior performance with the area under curve (AUC) of 0.90 in both models, alongside high sensitivity, specificity, and F1-score. This research highlights the potential of using a machine learning-based LightGBM model to facilitate early and accurate diagnosis of AL amyloidosis. The model's effectiveness suggests it could be a valuable tool in clinical settings, aiding in the timely identification of AL amyloidosis among patients with non-specific symptoms. Further validation in diverse populations is recommended to establish its universal applicability.

Pore and fracture pressures prediction — A new geomechanic approach in deepwater salt overthrusts: Case histories from the Gulf of Mexico

Along this active exploration belt, applying the conventional effective stress methods and algorithms, wherein the maximum stress ([Formula: see text]) is vertical and can be attained from the subsurface overburden (OB), would lead to unintended and unrealistic results. In the frontier active thrust belt of the Gulf of Mexico, the unique geomechanical setting of [Formula: see text] as the lateral salt creep and the minimum stress ([Formula: see text]) as the OB greatly impact the formation geopressure framework. The buoyancy of thick salt produces two different pressure gradients (PGs) above and below the salt. Moreover, the inclusion of rafted sediments in the salt and the plowing rubble zone at the salt base substantially affects the pore and fracture pressures (PP-FP) profiles. These proceeding geologic settings are the foundation for the conceptual framework. Building an alternative predrilling prediction numerical model based on these anomalous geomechanical settings and the lack of adequate seismic velocity is a challenge. All the available direct measured or pertained PP-FP data from the key wells are collected and tabulated. Prediction models are established by correlating the populated database and generating the empirical algorithm for each data gather. A substantial discrepancy occurs between above and below the salt sections where high PG in the sediment above the salt and slow PG development below the salt. There is considerable regressive pressure (average 2 ppg) in PP-FP subsalt sections. The PP within the salt is contingent on the presence of sediment inclusions, and a substantial FP drop in the rubble zone leads to the extensive loss of mud circulation. The trend lines of each data gather led to generating two depth-dependent equations for the PP-FP above and within the salt and two others for the subsalt. The prediction models are validated against a blind data set. Before drilling, this model establishes the PP-FP versus the sediment subsea depth in an abnormal geomechanical setting and the lack of coherent vertical seismic velocity for PP-FP predictions.

Modelling multivariate spatio-temporal data with identifiable variational autoencoders

Modelling multivariate spatio-temporal data with complex dependency structures is a challenging task but can be simplified by assuming that the original variables are generated from independent latent components. If these components are found, they can be modelled univariately. Blind source separation aims to recover the latent components by estimating the unknown linear or nonlinear unmixing transformation based on the observed data only. In this paper, we extend recently introduced identifiable variational autoencoder to the nonlinear nonstationary spatio-temporal blind source separation setting and demonstrate its performance using comprehensive simulation studies. Additionally, we introduce two alternative methods for the latent dimension estimation, which is a crucial task in order to obtain the correct latent representation. Finally, we illustrate the proposed methods using a meteorological application, where we estimate the latent dimension and the latent components, interpret the components, and show how nonstationarity can be accounted and prediction accuracy can be improved by using the proposed nonlinear blind source separation method as a preprocessing method.

UV-light fluorescence as a confirmation method for presumptive Legionella colonies isolated from water samples.

Legionnaires' disease (LD) is a severe and potentially fatal form of bacterial pneumonia caused by Legionella spp. We evaluated the use of UV-light for detecting Legionella non-pneumophila in water samples according to the NEN-EN-ISO 11731:2017 methodology (reference method) in a collaborative effort involving 10 laboratories. First, a test panel was constructed of 298 strains: 157 Legionella strains and 141 non-Legionella strains were cultured on buffered charcoal yeast extract (BCYE)-medium and confirmed according to ISO 11731: 2017 (cultured on BCYE agar plates with and without l-cysteine), and by matrix-assisted laser desorption-ionization time of flight or next generation sequencing. All strains were additionally exposed to an UV-light to assess if they showed a bright blue fluorescence effect (UV-positive) or not (UV-negative). Second, in an interlaboratory study, 10 laboratories analyzed a blinded set of 16 Legionella strains and 8 non-Legionella strains using both methods. The test panel analyses showed 100% accordance between the UV-light and reference method. In addition, the interlaboratory study results showed full agreement between both methods. Our results support the implementation of UV-light detection to confirm Legionella presumptive colonies during analyses of water samples according to the NEN-EN-ISO 11731:2017 methodology. Implementation of UV-light confirmation could reduce workload, time-to-result and costs for the analyses of water samples for the presence of Legionella.

Improving Specificity for Ovarian Cancer Screening Using a Novel Extracellular Vesicle–Based Blood Test: Performance in a Training and Verification Cohort

The low incidence of ovarian cancer (OC) dictates that any screening strategy needs to be both highly sensitive and highly specific. This study explored the utility of detecting multiple colocalized proteins or glycosylation epitopes on single tumor-associated extracellular vesicles (EVs) from blood. The novel OC Test employs immunoaffinity capture of tumor-associated EVs followed by proximity-ligation qPCR to detect combinations of up to three biomarkers to maximize specificity and measures multiple combinations to maximize sensitivity. A high-grade serous carcinoma (HGSC) case-control training set of EDTA plasma samples from 397 women was used to lock down the test design, the data interpretation algorithm, and the cut-off between cancer and non-cancer. Performance was verified and compared to CA125 in an independent blinded case-control set of serum samples from 390 women (132 controls, 66 HGSC, 83 non-HGSC OC, 109 benign). In the verification study, the OC Test showed a specificity of 97.0% (128/132; 95% CI: 92.4%-99.6%), a HGSC sensitivity of 97.0% (64/66; 95% CI: 87.8%-99.2%), and an AUC of 0.97 (95% CI 0.93-0.99) and also detected 73.5% (61/83; 95% CI: 62.7%-82.6%) of the non-HGSC OC cases. This test exhibited fewer false positives in subjects with benign ovarian tumors, non-ovarian cancers, and inflammatory conditions when compared to CA125. The combined sensitivity and specificity of this new test suggests it may have potential in OC screening.

Immunojanus Particles for low-volume and isolation-free unlabeled characterization of small Extracellular Vesicle in biofluids: Characterization of disease type by surface marker profiling.

Small extracellular vesicles (sEVs) are vital for cellular communication and serve as critical biomarker carriers for diseases such as cancer. However, quantifying and profiling sEV surface markers presents significant challenges due to the low concentration of specific sEV-bound proteins and interference by more abundant dispersed proteins. This paper presents Immunojanus Particles (IJPs), a new method that enables the direct detection of sEVs in less than an hour without isolation. The design of IJPs incorporates fluorescent and non-fluorescent halves, utilizing rotational Brownian motion to detect captured sEVs through the change in the blinking rate, without interference from the smaller dispersed proteins. We demonstrate a detection limit of 2E5 sEVs/mL with low sample volumes and the capability to characterize sEVs directly from plasma, serum, cell culture media, and urine. In a small pilot study involving 87 subjects, including individuals with colorectal cancer, pancreatic ductal adenocarcinoma, glioblastoma, Alzheimer's disease, and healthy controls, our method accurately identified the type of disease with high 0.90-0.99 AUC in a blind setting. Compared with an orthogonal ultracentrifugation plus surface plasmon resonance (UC+SPR) method that requires about 24 hours, the sensitivity and dynamic range of IJP are better by 2 logs.

Speed and efficiency: evaluating pulmonary nodule detection with AI-enhanced 3D gradient echo imaging.

Evaluating the diagnostic feasibility of accelerated pulmonary MR imaging for detection and characterisation of pulmonary nodules with artificial intelligence-aided compressed sensing. In this prospective trial, patients with benign and malignant lung nodules admitted between December 2021 and December 2022 underwent chest CT and pulmonary MRI. Pulmonary MRI used a respiratory-gated 3D gradient echo sequence, accelerated with a combination of parallel imaging, compressed sensing, and deep learning image reconstruction with three different acceleration factors (CS-AI-7, CS-AI-10, and CS-AI-15). Two readers evaluated image quality (5-point Likert scale), nodule detection and characterisation (size and morphology) of all sequences compared to CT in a blinded setting. Reader agreement was determined using the intraclass correlation coefficient (ICC). Thirty-seven patients with 64 pulmonary nodules (solid n = 57 [3-107 mm] part-solid n = 6 [ground glass/solid 8 mm/4-28 mm/16 mm] ground glass nodule n = 1 [20 mm]) were analysed. Nominal scan times were CS-AI-7 3:53 min; CS-AI-10 2:34 min; CS-AI-15 1:50 min. CS-AI-7 showed higher image quality, while quality remained diagnostic even for CS-AI-15. Detection rates of pulmonary nodules were 100%, 98.4%, and 96.8% for CS-AI factors 7, 10, and 15, respectively. Nodule morphology was best at the lowest acceleration and was inferior to CT in only 5% of cases, compared to 10% for CS-AI-10 and 23% for CS-AI-15. The nodule size was comparable for all sequences and deviated on average < 1 mm from the CT size. The combination of compressed sensing and AI enables a substantial reduction in the scan time of lung MRI while maintaining a high detection rate of pulmonary nodules. Incorporating compressed sensing and AI in pulmonary MRI achieves significant time savings without compromising nodule detection or characteristics. This advancement holds clinical promise, enhancing efficiency in lung cancer screening without sacrificing diagnostic quality. Lung cancer screening by MRI may be possible but would benefit from scan time optimisation. Significant scan time reduction, high detection rates, and preserved nodule characteristics were achieved across different acceleration factors. Integrating compressed sensing and AI in pulmonary MRI offers efficient lung cancer screening without compromising diagnostic quality.

Comparison of radiography and computed tomography for identification of third metacarpal structural change and associated assessment of condylar stress fracture risk in Thoroughbred racehorses.

Catastrophic injury has a low incidence but leads to the death of many Thoroughbred racehorses. To determine sensitivity, specificity, and reliability for third metacarpal condylar stress fracture risk assessment from digital radiographs (DR) and standing computed tomography (sCT). Controlled ex vivo experiment. A blinded set of metacarpophalangeal joint DR and sCT images were prepared from 31 Thoroughbreds. Four observers evaluated the condyles and parasagittal grooves (PSG) of the third metacarpal bone for the extent of dense bone and lucency/fissure and assigned a risk assessment grade for condylar stress fracture based on imaging features. Sensitivity and specificity for detection of subchondral structural changes in the condyles and PSG, and for risk assessment for condylar stress fracture were determined by comparison with a reference assessment based on sCT and joint surface examination. Agreement between observers and the reference assessment and reliability between observers were determined. Intra-observer repeatability was also assessed. Sensitivity for detection of structural change was lower than specificity for both imaging methods and all observers. For agreement with the reference assessment of structural change, correlation coefficients were generally below 0.5 for DR and 0.49-0.82 for sCT. For horses categorised as normal risk on reference assessment, observer assessment often agreed with the reference. Sensitivity for risk assessment was lower than specificity for all observers. For horses with a reference assessment of high risk of injury, observers generally underestimated risk. Diagnostic sensitivity of risk assessment was improved with sCT imaging, particularly for horses categorised as having elevated risk of injury from the reference assessment. Assessment repeatability and reliability was better with sCT than DR. The ex vivo study design influenced DR image sets. Risk assessment through screening with diagnostic imaging is a promising approach to improve injury prevention in racing Thoroughbreds. Knowledge of sensitivity and specificity of fetlock lesion detection provides the critical guidance needed to improve racehorse screening programs. We found improved detection of MC3 subchondral structural change and risk assessment for condylar stress fracture with sCT ex vivo.

GPU accelerated sparse curvelet-constrained wavefield reconstruction inversion with source estimation: Application to Chevron Benchmark 2014 blind test data set

Full-waveform inversion (FWI) is a pivotal tool in seismic exploration and seismology but encounters a persistent challenge known as “cycle-skipping,” causing it to frequently converge to local minima. Wavefield reconstruction inversion (WRI) has emerged as a potential solution to mitigate the risk of cycle skipping. Researchers have provided numerical examples indicating that WRI is less prone to being ensnared in local minima caused by the absence of low-frequency component data and the absence of a well-defined initial model compared with conventional FWI. Despite its potential, the computational demands of WRI have hindered its widespread application, especially in scenarios involving ocean towed streamer seismic acquisition and unknown sources, where the augmented systems differ from source to source. In our study, we introduce a novel approach — sparse curvelet-constrained WRI with source estimation (WRI-SE-CC) — accelerated by graphics processing unit (GPU). Real-time source function estimation is achieved through the variable projection method, and noise-related artifacts are suppressed using sparse curvelet constraints. By optimizing the utilization of hundreds of computation processors within a GPU for parallel computing of matrix-vector multiplications, we present a GPU-based grouped conjugate gradient method to accelerate the computation of WRI-SE-CC. Numerical experiments demonstrate a significant 240-fold acceleration compared with the preconditioned conjugate gradient using one CPU core for computations involving multiple sources. Inversion experiments with the overthrust model demonstrate the capability of our method in mitigating local minima and suppressing noise-related artifacts. Furthermore, we validate the framework on the Chevron 2014 blind test data set, showcasing its effectiveness in addressing practical challenges in the field.

Novel approaches in geomechanical parameter estimation using machine learning methods and conventional well logs

ABSTRACT Today, geomechanics plays a crucial role in the oil industry, particularly in enhancing production and ensuring well stability. To achieve optimal results, accurate estimation of geomechanical parameters is essential. One of the low-cost and accurate methods for estimating geomechanical parameters is the use of intelligent methods. In this research, geomechanical parameters are estimated using conventional data logs using intelligent methods. The aim of this study is to introduce a new machine learning algorithm to estimate geomechanical parameters using conventional data logs in one of the hydrocarbon field wells in southwest Iran. In this article, the shear wave velocity and uniaxial compressive strength (UCS) were estimated using machine learning algorithms. Subsequently, other geomechanical parameters were calculated based on these estimated parameters derived from machine learning algorithms. For shear wave velocity (Vs) prediction using MLP and CLM (CNN+LSTM+MLP) algorithms, First, effective features were selected using the Auto-encoder deep learning algorithm. The selected features for Vs input into the algorithms were Vp, RHOB, CALIPER, and NPHI, and then the Vs is estimated with MLP and CLM algorithm. To evaluate the results, the model was assessed using metrics such as MAE, MAPE, MSE, RMSE, NRMSE, and R2 on the train, test, and blind datasets. The CLM algorithm consistently demonstrated superior performance across all datasets, including training, testing, and blind data sets. The R2 values for blind data were R M L P 2 = 0.8727 , R C L M 2 = 0.9274 , respectively. These outputs are crucial for estimating subsequent studies. Next, Elastic Young’s moduli and Poisson’s ratio were calculated, and the dynamic brittleness index was computed using dynamic Young’s modulus and Poisson’s ratio. Subsequently, UCS values were predicted using machine learning algorithms. Since there were 12 laboratory core samples of UCS available, UCS was initially calculated using relevant empirical relations and data from some available well logs. This process aimed to extrapolate these core results to cover the entire target depth range from 3551.072 to 3799.789 m. Subsequently, a relationship between UCS derived from well logs and laboratory results was established for the depths corresponding to where the laboratory samples were recorded. Following that, an autoencoder deep network was utilized to select effective features for predicting UCS. The selected features for UCS input into the algorithms were Vp, RHOB, and CALIPER. Subsequently, UCS was estimated using MLP and CLM algorithms. To evaluate the results, the model performance was assessed using metrics such as MAE, MAPE, MSE, RMSE, NRMSE, and R2 on the train, test, and blind datasets. Furthermore, the results were checked and evaluated using a set of statistical measures calculated for the train, test, and blind datasets, that R M L P 2 = 0.9305 , R C L M 2 = 0.9953 were obtained for blind UCS data. The results demonstrate that CLM achieves high accuracy in estimating these parameters, with deep learning algorithms showing higher determination coefficients and lower errors compared to MLP. Furthermore, UCS and tensile strength were calculated, followed by the computation of static brittleness index. The relationship between dynamic and static brittleness index was also investigated. Overall, the findings indicate that machine learning algorithms are robust and accurate methods for estimating Vs, UCS, and other geomechanical parameters using conventional logs.

P-256 AI-enhanced precision: unveiling the future of blastocyst quality assessment

Abstract Study question Can artificial intelligence (AI) effectively assess various regions of embryos to estimate the quality of human embryos captured through an optical microscope? Summary answer Artificial Intelligence excels in analyzing diverse embryonic regions based on morphology, ensuring precise grade predictions, assessing human embryos via optical microscopes. What is known already In recent research, the integration of AI and computer vision has marked significant strides in embryo selection, providing precise predictions of clinical pregnancy based on images of human embryos, particularly at Day 5. This study extends its findings by incorporating techniques that exhibit resilience to variations in camera and microscope types. Notably, the AI algorithms underwent rigorous testing on double-blind datasets, comprising optical microscopic images sourced from Optical microscope systems across diverse regions such as India, Europe, the UK, the US, and Australia. These results signify a promising advancement in the standardization and applicability of AI-assisted embryo assessment methodologies. Study design, size, duration Approximately 11824 static 2D images of Day 5 blastocysts with related pregnancy, preimplantation genetic testing for aneuploidy (PGT-A) outcomes, demographic and clinic geographical location information have been collected. Images were divided into three groups: training, validation and blind test sets. An AI model was trained on 3473 embryo images, validated on 1184 and tested on further blind set of 1253 images from a separate clinics and demographsic. Gardner based grading system was used. Participants/materials, setting, methods Dataset included optical-microscope images from fertility patients across 25 IVF laboratories in six countries was utilized to train, validate, and test the AI. Dataset was graded by 9 embryologists and final grade was decided based on max-voting. The identity of participants was kept secret and did not associate with used data. Assessment encompassed accuracy, distribution, and robustness concerning camera/microscope type. The dataset spanned various embryo development stages, excluding frozen, degenerated, non-blastocyst embryos. Main results and the role of chance The study distinguishes itself by incorporating data from diverse environments, enhancing generalizability. Leveraging transformer-based attention models ensures meticulous focus on regions relevant to the Gardner grading system, ensuring accurate predictions akin to human judgment and minimizing errors prone to subjectivity. AI’s predictive accuracy on blind dataset for expansion is 82.18%, for ICM is 76.53%, and for trophectoderm is 77.53%, surpassing previous studies. The model demonstrates specificity 79.16%, sensitivity of 85.95% and accuracy of 82.44% in estimating embryo quality. Direct application on embryoscope images is hindered by differing morphology, excluding the embryo part. However, employing segmentation and applying models yields comparable performance with approx. 2.87% drop in accuracy. These findings suggest that the proposed method, with AI trained on a globally diverse dataset, establishes a generalizable AI robust to variations in camera type and focal settings. Limitations, reasons for caution Data collected for this solution has been limited to mainly static images. The study’s focus on good-quality Day 5 blastocyst embryos might limit generalizability to other stages. The dataset’s global diversity may not fully capture demographic nuances. Ongoing AI advancements may necessitate continual updates for relevance and accuracy. Wider implications of the findings Broader implications of the findings suggests transformative potential for AI in embryo quality assessment. The methodology’s success in diverse environments and robustness to variations in camera types underscore its applicability on global scale. This could lead to standardized, automated, and precise embryo evaluations, offering significant advancements in assisted reproductive technology. Trial registration number not applicable

Individualized epidemic spreading models predict epilepsy surgery outcomes: A pseudo-prospective study.

Epilepsy surgery is the treatment of choice for drug-resistant epilepsy patients, but up to 50% of patients continue to have seizures one year after the resection. In order to aid presurgical planning and predict postsurgical outcome on a patient-by-patient basis, we developed a framework of individualized computational models that combines epidemic spreading with patient-specific connectivity and epileptogeneity maps: the Epidemic Spreading Seizure and Epilepsy Surgery framework (ESSES). ESSES parameters were fitted in a retrospective study (N = 15) to reproduce invasive electroencephalography (iEEG)-recorded seizures. ESSES reproduced the iEEG-recorded seizures, and significantly better so for patients with good (seizure-free, SF) than bad (nonseizure-free, NSF) outcome. We illustrate here the clinical applicability of ESSES with a pseudo-prospective study (N = 34) with a blind setting (to the resection strategy and surgical outcome) that emulated presurgical conditions. By setting the model parameters in the retrospective study, ESSES could be applied also to patients without iEEG data. ESSES could predict the chances of good outcome after any resection by finding patient-specific model-based optimal resection strategies, which we found to be smaller for SF than NSF patients, suggesting an intrinsic difference in the network organization or presurgical evaluation results of NSF patients. The actual surgical plan overlapped more with the model-based optimal resection, and had a larger effect in decreasing modeled seizure propagation, for SF patients than for NSF patients. Overall, ESSES could correctly predict 75% of NSF and 80.8% of SF cases pseudo-prospectively. Our results show that individualised computational models may inform surgical planning by suggesting alternative resections and providing information on the likelihood of a good outcome after a proposed resection. This is the first time that such a model is validated with a fully independent cohort and without the need for iEEG recordings.

Blind prediction of fracture in an additively manufactured geomaterial using the finite discrete element method

Modeling fracture and damage processes is a challenging, yet critical capability for promoting safe and sustainable design in geotechnical, energy, and hydrological applications. Numerical methods capable of modeling fracture processes are rarely asked to make blind predictions against experimental observations. Here, we report results from a numerical study of blind fracture prediction in an additively manufactured, three-point bend specimen. We record a number of physical mechanisms associated with the fracture of the specimen and make a blind prediction against experimental results that were unknown when the numerical study was performed. We then compare results from our model to the blind data set. We find that the combined finite discrete element method performs exceptionally well in the simulation and prediction of fracture processes.

Application of a Machine Learning-Based Classification Approach for Developing Host Protein Diagnostic Models for Infectious Disease

In recent years, infectious disease diagnosis has increasingly turned to host-centered approaches as a complement to pathogen-directed ones. The former, however, typically requires the interpretation of complex multiple biomarker datasets to arrive at an informative diagnostic outcome. This report describes a machine learning (ML)-based classification workflow that is intended as a template for researchers seeking to apply ML approaches for developing host-based infectious disease biomarker classifiers. As an example, we built a classification model that could accurately distinguish between three disease etiology classes: bacterial, viral, and normal in human sera using host protein biomarkers of known diagnostic utility. After collecting protein data from known disease samples, we trained a series of increasingly complex Auto-ML models until arriving at an optimized classifier that could differentiate viral, bacterial, and non-disease samples. Even when limited to a relatively small training set size, the model had robust diagnostic characteristics and performed well when faced with a blinded sample set. We present here a flexible approach for applying an Auto-ML-based workflow for the identification of host biomarker classifiers with diagnostic utility for infectious disease, and which can readily be adapted for multiple biomarker classes and disease states.

Zero-shot denoising of microscopy images recorded at high-resolution limits.

Conventional and electron microscopy visualize structures in the micrometer to nanometer range, and such visualizations contribute decisively to our understanding of biological processes. Due to different factors in recording processes, microscopy images are subject to noise. Especially at their respective resolution limits, a high degree of noise can negatively effect both image interpretation by experts and further automated processing. However, the deteriorating effects of strong noise can be alleviated to a large extend by image enhancement algorithms. Because of the inherent high noise, a requirement for such algorithms is their applicability directly to noisy images or, in the extreme case, to just a single noisy image without a priori noise level information (referred to as blind zero-shot setting). This work investigates blind zero-shot algorithms for microscopy image denoising. The strategies for denoising applied by the investigated approaches include: filtering methods, recent feed-forward neural networks which were amended to be trainable on noisy images, and recent probabilistic generative models. As datasets we consider transmission electron microscopy images including images of SARS-CoV-2 viruses and fluorescence microscopy images. A natural goal of denoising algorithms is to simultaneously reduce noise while preserving the original image features, e.g., the sharpness of structures. However, in practice, a tradeoff between both aspects often has to be found. Our performance evaluations, therefore, focus not only on noise removal but set noise removal in relation to a metric which is instructive about sharpness. For all considered approaches, we numerically investigate their performance, report their denoising/sharpness tradeoff on different images, and discuss future developments. We observe that, depending on the data, the different algorithms can provide significant advantages or disadvantages in terms of their noise removal vs. sharpness preservation capabilities, which may be very relevant for different virological applications, e.g., virological analysis or image segmentation.

Prism-shaped rock-physics template

Despite the emergence of statistical and intelligent approaches for quantitative seismic interpretation in recent years, rock-physics templates (RPTs) are still preferred because of their simplicity and easy implementation. RPTs have been introduced as a fundamental tool for lithology and fluid discrimination based on well-log and seismic data, which have already been proven in terms of accuracy and reliability. Considering the demand for comprehensive RPTs and improving their efficiency, I develop a novel 3D prism-shaped template for lithology and fluid discrimination called a prism-shaped RPT (P-RPT). For this, a theoretical methodology for designing conventional RPT is introduced with the aim of bounds and hybrid rock-physics models. Next, some novel triangular and rectangular ternary and binary templates are designed and connected to one another to build a prism for fluid discrimination and lithology identification. Two ternary charts are used for fluid and lithology, incorporating the P- and S-wave velocity ratios and lambda-mu parameters. Furthermore, binary charts, including acoustic impedance, Lamé parameters, and porosity, are designed and modified from the literature for fluid and lithology identification. Next, the obtained templates are successfully validated on blind data sets (ultrasonic, well logging, and seismic data) in different reservoirs with various lithologies and fluid types. The results indicate that the P-RPT could integrate the available RPTs into a 3D diagram and give a reliable framework for applications in seismic interpretation. User friendliness and generalizability are the most prominent advantages of this template for detecting fluid type and lithology based on well logs and seismic inversion, the results of which can be interpreted on a single chart rather than analyzing the data with different templates. The methodology and framework for implementing and generating templates are thoroughly explained in theory and practice to localize P-RPT or update the new RPT for another region.

Validation of Patient-Reported Outcomes in Patients With Nonmetastatic Breast Cancer Receiving Comprehensive Nodal Irradiation in the RadComp Trial

Our purpose was to evaluate the measurement properties of patient-reported outcome (PRO) measures used in the ongoing RadComp pragmatic randomized clinical trial (PRCT). The deidentified and blinded data set included 774 English-speaking female participants who completed their 6-month posttreatment assessment. Eleven PRO measures were evaluated, including the Trial Outcome Index from the Functional Assessment of Cancer Therapy-Breast (FACT-B), Satisfaction with Breast Cosmetic Outcomes, the BREAST-Q, and selected Patient-Reported Outcomes Measurement Information System (PROMIS) measures. PROs were measured at 3 timepoints: baseline, completion of radiation therapy (RT), and 6 months post-RT. Ten variables were used as validity anchors. Pearson or Spearman correlations were calculated between PROs and convergent validity indicators. Mean PRO differences between clinically distinct categories were compared with analysis of variance methods (known-groups validity). PRO change scores were mapped to change in other variables (sensitivity to change). Most correlations between PROs and validity indicators were large (≥0.5). Mean score for Satisfaction with Breast Cosmetic Outcomes was higher (better) for those with a lumpectomy compared with those with a mastectomy (P < .001). Mean scores for the FACT-B Trial Outcome Index and for PROMIS Fatigue and Ability to Participate in Social Roles and Activities were better for those with good baseline performance status compared with those with poorer baseline performance status (P < .05). At completion of RT and post-RT, mean scores for Satisfaction with Breast Cosmetic Outcomes and BREAST-Q Radiation were significantly different (P < .001) across categories for all Functional Assessment of Chronic Illness Therapy -Treatment Satisfaction - General items. There were medium-sized correlations between change scores for FACT-B Trial Outcome Index, Fatigue, Anxiety, and Ability to Participate in Social Roles and change scores in the Visual Analog Scale. For patients with nonmetastatic breast cancer receiving radiation in the RadComp PRCT, our findings demonstrate high reliability and validity for important PRO measures, supporting their psychometric strength and usefulness to reflect the effect of RT on health-related quality of life.