Calibrating Deep Learning Classifiers for Patient-Independent Electroencephalogram Seizure Forecasting.

The recent scientific literature abounds in proposals of seizure forecasting methods that exploit machine learning to automatically analyze electroencephalogram (EEG) signals. Deep learning algorithms seem to achieve a particularly remarkable performance, suggesting that the implementation of clinical devices for seizure prediction might be within reach. However, most of the research evaluated the robustness of automatic forecasting methods through randomized cross-validation techniques, while clinical applications require much more stringent validation based on patient-independent testing. In this study, we show that automatic seizure forecasting can be performed, to some extent, even on independent patients who have never been seen during the training phase, thanks to the implementation of a simple calibration pipeline that can fine-tune deep learning models, even on a single epileptic event recorded from a new patient. We evaluate our calibration procedure using two datasets containing EEG signals recorded from a large cohort of epileptic subjects, demonstrating that the forecast accuracy of deep learning methods can increase on average by more than 20%, and that performance improves systematically in all independent patients. We further show that our calibration procedure works best for deep learning models, but can also be successfully applied to machine learning algorithms based on engineered signal features. Although our method still requires at least one epileptic event per patient to calibrate the forecasting model, we conclude that focusing on realistic validation methods allows to more reliably compare different machine learning approaches for seizure prediction, enabling the implementation of robust and effective forecasting systems that can be used in daily healthcare practice.

Comparative near Infrared (NIR) spectroscopy calibrations performance of dried and undried forage on dry and wet matter bases.

The application of Near Infrared (NIR) spectroscopy for analyzing wet feed directly on farms is increasingly recognized for its role in supporting harvest-time decisions and refining the precision of animal feeding practices. This study aims to evaluate the accuracy of NIR spectroscopy calibrations for both undried, unprocessed samples and dried, ground samples. Additionally, it investigates the influence of the bases of reference data (wet vs. dry basis) on the predictive capabilities of the NIR analysis. The study utilized 492 Corn Whole Plant (CWP) and 405 High Moisture Corn (HMC) samples, sourced from various farms across Italy. Spectral data were acquired from both undried, unground and dried, ground samples using laboratory bench NIR instruments, covering a spectral range of 1100 to 2498 nm. The reference chemical composition of these samples was analyzed and presented in two formats: on a wet matter basis and on a dry matter basis. The study revealed that calibrations based on undried samples generally exhibited lower predictive accuracy for most traits, with the exception of Dry Matter (DM). Notably, the decline in predictive performance was more pronounced in highly moist products like CWP, where the average error increased by 60-70%. Conversely, this reduction in accuracy was relatively contained (10-15%) in drier samples such as HMC. The Standard Error of Cross-Validation (SECV) values for DMres, Ash, CP, and EE were notably low, at 0.39, 0.30, 0.29, 0.21% for CWP and 0.49, 0.14, 0.25, 0.14% for HMC, respectively. These results align with previous studies, indicating the reliability of NIR spectroscopy in diverse moisture contexts. The study attributes this variance to the interference caused by water in 'as is' samples, where the spectral features predominantly reflect water content, thereby obscuring the spectral signatures of other nutrients. In terms of calibration development strategies, the study concludes that there is no significant difference in predictive performance between undried calibrations based on either 'dry matter' or 'as is' basis. This finding emphasizes the potential of NIR spectroscopy in diverse moisture contexts, although with varying degrees of accuracy contingent upon the moisture content of the analyzed samples. Overall, this research provides valuable insights into the calibration strategies of NIR spectroscopy and its practical applications in agricultural settings, particularly for on-farm forage analysis.

RFPID: development and 3D-printing of a female physical phantom for whole-body counter.

Whole-body counters (WBC) are used in internal dosimetry forin vivomonitoring in radiation protection. The calibration processes of a WBC set-up include the measurement of a physical phantom filled with a certificate radioactive source that usually is referred to a standard set of individuals determined by the International Commission on Radiological Protection (ICRP). The aim of this study was to develop an anthropomorphic and anthropometric female physical phantom for the calibration of the WBC systems. The reference female computational phantom of the ICRP, now called RFPID (Reference Female Phantom for Internal Dosimetry) was printed using PLA filament and with an empty interior. The goal is to use the RFPID to reduce the uncertainties associated within vivomonitoring system. The images which generated the phantom were manipulated using ImageJ®, Amide®, GIMP®and the 3D Slicer®software. RFPID was split into several parts and printed using a 3D printer in order to print the whole-body phantom. The newly printed physical phantom RFPID was successfully fabricated, and it is suitable to mimic human tissue, anatomically similar to a human body i.e., size, shape, material composition, and density.

Usability of sand samples in TL Radiation Dosimetry.

This study investigated natural sand thermoluminescence (TL) response as a possible option for retrospective high-dose gamma dosimetry. The natural sand under investigation was collected from six locations with selection criteria for sampling sites covering the highest probability of exposure to unexpected radiation on the Egyptian coast. Dose-response, glow curve, chemical composition, linearity, and fading rate for different sand samples were studied. Energy Dispersive X-ray Spectroscopy (EDX) analysis revealed differences in chemical composition among the various geological sites, leading to variations in TL glow curve intensity. Sand samples collected from Ras Sedr, Taba, Suez, and Enshas showed similar TL patterns, although with different TL intensities. Beach sands of Matrouh and North Coastal with a high calcite content did not show a clear linear response to the TL technique, in the dose range of 10 Gy up to 30 kGy. The results show that most sand samples are suitable as a radiation dosimeter at accidental levels of exposure. It is proposed here that for high-dose gamma dosimetry with doses ranging from 3 to 10 kGy, a single calibration factor might be enough for TL measurements using sand samples. However, proper calibration might allow dose assessment for doses even up to 30 kGy. Most of the investigated sand samples had nearly stable fading rates after seven days of storage. The Ras Sedr sands sample was the most reliable for retrospective dose reconstruction.

Calibration of commercial fisheries echo sounders using seabed backscatter for the estimation of fishery resources.

Acoustic methods are often used for fisheries resource surveys to investigate fish stocks in a wide area. Commercial fisheries echo sounders, which are installed on most small fishing vessels, are used to record a large amount of data during fishing trips. Therefore, it can be used to collect the basic information necessary for stock assessment for a wide area and frequently. To carry out the quantification for the fisheries echo sounder, we devised a simple method using the backscattering strength of the seabed to perform calibration periodically and easily. In this study, seabed secondary reflections were used instead of primary reflection because the fisheries echo sounders were not equipped with a time-varied gain (TVG) function, and the primary backscattering strength of the seabed was saturated. It was also necessary to use standard values of seabed backscattering strength averaged over a certain area for calibration to eliminate some of the effects of differences in seabed sediment and vessel motions. By using standard values of the seabed secondary reflections, the fisheries echo sounder was calibrated accurately. Our study can provide a reliable framework to calibrate commercial fisheries echo sounders, to improve the estimation and management of fishery resources.

Development of a machine learning-based model to predict prognosis of alpha-fetoprotein-positive hepatocellular carcinoma.

BACKGROUND: Patients with alpha-fetoprotein (AFP)-positive hepatocellular carcinoma (HCC) have aggressive biological behavior and poor prognosis. Therefore, survival time is one of the greatest concerns for patients with AFP-positive HCC. This study aimed to demonstrate the utilization of six machine learning (ML)-based prognostic models to predict overall survival of patients with AFP-positive HCC. METHODS: Data on patients with AFP-positive HCC were extracted from the Surveillance, Epidemiology, and End Results database. Six ML algorithms (extreme gradient boosting [XGBoost], logistic regression [LR], support vector machine [SVM], random forest [RF], K-nearest neighbor [KNN], and decision tree [ID3]) were used to develop the prognostic models of patients with AFP-positive HCC at one year, three years, and five years. Area under the receiver operating characteristic curve (AUC), confusion matrix, calibration curves, and decision curve analysis (DCA) were used to evaluate the model. RESULTS: A total of 2,038 patients with AFP-positive HCC were included for analysis. The 1-, 3-, and 5-year overall survival rates were 60.7%, 28.9%, and 14.3%, respectively. Seventeen features regarding demographics and clinicopathology were included in six ML algorithms to generate a prognostic model. The XGBoost model showed the best performance in predicting survival at 1-year (train set: AUC = 0.771; test set: AUC = 0.782), 3-year (train set: AUC = 0.763; test set: AUC = 0.749) and 5-year (train set: AUC = 0.807; test set: AUC = 0.740). Furthermore, for 1-, 3-, and 5-year survival prediction, the accuracy in the training and test sets was 0.709 and 0.726, 0.721 and 0.726, and 0.778 and 0.784 for the XGBoost model, respectively. Calibration curves and DCA exhibited good predictive performance as well. CONCLUSIONS: The XGBoost model exhibited good predictive performance, which may provide physicians with an effective tool for early medical intervention and improve the survival of patients.

Development and validation of a nomogram model for predicting distant metastasis of aged ≥50 patients with thyroid carcinoma: a SEER database analysis.

OBJECTIVE: This work aimed to construct and validate a model for predicting distant metastasis (DM) in thyroid carcinoma (TC) patients aged≥50. PATIENTS AND METHODS: The research data were collected from the Surveillance, Epidemiology, and End Results (SEER) program databases via SEER*Stat software (https://seer.cancer.gov/). Logistics regression was used to screen the independent risk factors for TC patients. The nomogram was constructed and validated based on the logistics regression results for predicting DM occurrence in TC patients. Moreover, the characteristic curves (ROC) were used to assess the predictive performance. The decision analysis curve (DCA) and the calibration curve were used to test this nomogram's accuracy and discrimination. Additionally, we analyzed survival and risk scores in TC patients with metastasis using the Kaplan-Meier (KM) method. RESULTS: A total of 11,166 TC patients were divided into a training set and a validation set. The results showed that topography (T), lymph node metastasis (N), and (grade) G were crucial risk factors for predicting DM. ROC analysis showed that the model had a good discriminative ability both in the training and validation set. The DCA curve showed greater net benefits across a range of DM risks for the nomogram in the training and validation set. Survival analyses showed that the metastasis cases with low-risk scores have shown a poorer prognosis in this study, both in the training and validation set. CONCLUSIONS: The nomogram model had excellent predictive performance and net benefit for predicting DM of TC patients aged ≥50. The model can help doctors develop treatment plans for their patients.

Enhancing Robotic Perception through Synchronized Simulation and Physical Common-Sense Reasoning.

We introduce both conceptual and empirical findings arising from the amalgamation of a robotics cognitive architecture with an embedded physics simulator, aligning with the principles outlined in the intuitive physics literature. The employed robotic cognitive architecture, named CORTEX, leverages a highly efficient distributed working memory known as deep state representation. This working memory inherently encompasses a fundamental ontology, state persistency, geometric and logical relationships among elements, and tools for reading, updating, and reasoning about its contents. Our primary objective is to investigate the hypothesis that the integration of a physics simulator into the architecture streamlines the implementation of various functionalities that would otherwise necessitate extensive coding and debugging efforts. Furthermore, we categorize these enhanced functionalities into broad types based on the nature of the problems they address. These include addressing challenges related to occlusion, model-based perception, self-calibration, scene structural stability, and human activity interpretation. To demonstrate the outcomes of our experiments, we employ CoppeliaSim as the embedded simulator and both a Kinova Gen3 robotic arm and the Open-Manipulator-P as the real-world scenarios. Synchronization is maintained between the simulator and the stream of real events. Depending on the ongoing task, numerous queries are computed, and the results are projected into the working memory. Participating agents can then leverage this information to enhance overall performance.

Development of a Robust Sensor Calibration for a Commercially Available Rising Platemeter to Estimate Herbage Mass on Temperate Seminatural Pastures.

Rising platemeters are commonly used in Ireland and New Zealand for managing intensive pastures. To assess the applicability of a commercial rising platemeter operating with a microsonic sensor to estimate herbage mass with its own equation, the objectives were (i) to validate the original equation; (ii) to identify possible factors hampering its accuracy and precision; and (iii) to develop a new equation for heterogeneous swards. A comprehensive dataset (n = 1511) was compiled on the pastures of dairy farms. Compressed sward heights were measured by the rising platemeter. Herbage mass was harvested to determine reference herbage availability. The adequacy of estimating herbage mass was assessed using root mean squared error (RMSE) and mean bias. As the adequacy of the original equation was low, a new equation was developed using multiple regression models. The mean bias and the RMSE for the new equation were overall low with 201 kg dry matter/ha and 34.6%, but it tended to overestimate herbage availability at herbage mass < 500 kg dry matter/ha and underestimate it at >2500 kg dry matter/ha. Still, the newly developed equation for the microsonic sensor-based rising platemeter allows for accurate and precise estimation of available herbage mass on pastures.

Causal Selection of Covariates in Regression Calibration for Mismeasured Continuous Exposure.

Regression calibration as developed by Rosner, Spiegelman, and Willett is used to adjust the bias in effect estimates due to measurement error in continuous exposures. The method involves two models: a measurement error model relating the mismeasured exposure to the true (or gold-standard) exposure and an outcome model relating the mismeasured exposure to the outcome. However, no comprehensive guidance exists for determining which covariates should be included in each model. In this article, we investigate the selection of the minimal and most efficient covariate adjustment sets under a causal inference framework. We show that to address the measurement error, researchers must adjust for, in both measurement error and outcome models, any common causes (1) of true exposure and the outcome and (2) of measurement error and the outcome. We also show that adjusting for so-called prognostic variables that are independent of true exposure and measurement error in the outcome model, may increase efficiency, while adjusting for any covariates that are associated only with true exposure generally results in efficiency loss in realistic settings. We apply the proposed covariate selection approach to the Health Professional Follow-up Study dataset to study the effect of fiber intake on cardiovascular disease. Finally, we extend the originally proposed estimators to a nonparametric setting where effect modification by covariates is allowed.

A deep learning approach for fast muscle water T2 mapping with subject specific fat T2 calibration from multi-spin-echo acquisitions.

This work presents a deep learning approach for rapid and accurate muscle water T2 with subject-specific fat T2 calibration using multi-spin-echo acquisitions. This method addresses the computational limitations of conventional bi-component Extended Phase Graph fitting methods (nonlinear-least-squares and dictionary-based) by leveraging fully connected neural networks for fast processing with minimal computational resources. We validated the approach through in vivo experiments using two different MRI vendors. The results showed strong agreement of our deep learning approach with reference methods, summarized by Lin's concordance correlation coefficients ranging from 0.89 to 0.97. Further, the deep learning method achieved a significant computational time improvement, processing data 116 and 33 times faster than the nonlinear least squares and dictionary methods, respectively. In conclusion, the proposed approach demonstrated significant time and resource efficiency improvements over conventional methods while maintaining similar accuracy. This methodology makes the processing of water T2 data faster and easier for the user and will facilitate the utilization of the use of a quantitative water T2 map of muscle in clinical and research studies.

A standard calibration method based on a symmetric resistance network matrix for galvanic logging instruments.

An increasing number of measurement electrodes have been designed to satisfy the demand for high-resolution detection using galvanic logging technology in complex formations. The forward modeling response analysis of logging tools has important guiding significance in the design of galvanic logging tools. Based on a three-dimensional finite element numerical simulation method, we established a forward model of galvanic multi-electrodes in a complex formation. We also designed a symmetrical resistance network model of the formation with equivalent resistance between two electrodes. A symmetrical resistance network was derived using the balanced bridge method. The asymmetrical admittance matrix was extended to a symmetrical extended admittance matrix to realize a convenient calculation of the equivalent symmetrical resistance network in complex formations. Verification of the microcolumn-focused logging tool, with nine electrodes in a simulated standard well, and an evaluation of the degree of invasion in an actual oil well indicate that this calibration method can improve the measurement accuracy of galvanic logging instruments.

Large-scale assessment of physical activity in a population using high-resolution hip-worn accelerometry: the German National Cohort (NAKO).

Large population-based cohort studies utilizing device-based measures of physical activity are crucial to close important research gaps regarding the potential protective effects of physical activity on chronic diseases. The present study details the quality control processes and the derivation of physical activity metrics from 100 Hz accelerometer data collected in the German National Cohort (NAKO). During the 2014 to 2019 baseline assessment, a subsample of NAKO participants wore a triaxial ActiGraph accelerometer on their right hip for seven consecutive days. Auto-calibration, signal feature calculations including Euclidean Norm Minus One (ENMO) and Mean Amplitude Deviation (MAD), identification of non-wear time, and imputation, were conducted using the R package GGIR version 2.10-3. A total of 73,334 participants contributed data for accelerometry analysis, of whom 63,236 provided valid data. The average ENMO was 11.7 ± 3.7 mg (milli gravitational acceleration) and the average MAD was 19.9 ± 6.1 mg. Notably, acceleration summary metrics were higher in men than women and diminished with increasing age. Work generated in the present study will facilitate harmonized analysis, reproducibility, and utilization of NAKO accelerometry data. The NAKO accelerometry dataset represents a valuable asset for physical activity research and will be accessible through a specified application process.

Thermoluminescent dosimeters (TLD-100) for absorbed dose measurements in alpha-emitting radionuclides.

Early works that used thermoluminescent dosimeters (TLDs) to measure absorbed dose from alpha particles reported relatively high variation (10%) between TLDs, which is undesirable for modern dosimetry applications. This work outlines a method to increase precision for absorbed dose measured using TLDs with alpha-emitting radionuclides by applying an alpha-specific chip factor (CF) that individually characterizes the TLD sensitivity to alpha particles. Variation between TLDs was reduced from 21.8% to 6.7% for the standard TLD chips and 7.9% to 3.3% for the thin TLD chips. It has been demonstrated by this work that TLD-100 can be calibrated to precisely measure the absorbed dose to water from alpha-emitting radionuclides.

[Research on Position Verification of Multi-Leaf Collimator (MLC) and Dose Verification Based on Electronic Portal Imaging Device].

Objective: A quality control (QC) system based on the electronic portal imaging device (EPID) system was used to realize the Multi-Leaf Collimator (MLC) position verification and dose verification functions on Primus and VenusX accelerators. Methods: The MLC positions were calculated by the maximum gradient method of gray values to evaluate the deviation. The dose of images acquired by EPID were reconstructed using the algorithm combining dose calibration and dose calculation. The dose data obtained by EPID and two-dimensional matrix (MapCheck/PTW) were compared with the dose calculated by Pinnacle/TiGRT TPS for γ passing rate analysis. Results: The position error of VenusX MLC was less than 1 mm. The position error of Primus MLC was significantly reduced after being recalibrated under the instructions of EPID. For the dose reconstructed by EPID, the average γ passing rates of Primus were 98.86% and 91.39% under the criteria of 3%/3 mm, 10% threshold and 2%/2 mm, 10% threshold, respectively. The average γ passing rates of VenusX were 98.49% and 91.11%, respectively. Conclusion: The EPID-based accelerator quality control system can improve the efficiency of accelerator quality control and reduce the workload of physicists.

Multi-Stimulus Least-Squares Transformation With Online Adaptation Scheme to Reduce Calibration Effort for SSVEP-Based BCIs.

Steady-state visual evoked potential (SSVEP), one of the most popular electroencephalography (EEG)-based brain-computer interface (BCI) paradigms, can achieve high performance using calibration-based recognition algorithms. As calibration-based recognition algorithms are time-consuming to collect calibration data, the least-squares transformation (LST) has been used to reduce the calibration effort for SSVEP-based BCI. However, the transformation matrices constructed by current LST methods are not precise enough, resulting in large differences between the transformed data and the real data of the target subject. This ultimately leads to the constructed spatial filters and reference templates not being effective enough. To address these issues, this paper proposes multi-stimulus LST with online adaptation scheme (ms-LST-OA). METHODS: The proposed ms-LST-OA consists of two parts. Firstly, to improve the precision of the transformation matrices, we propose the multi-stimulus LST (ms-LST) using cross-stimulus learning scheme as the cross-subject data transformation method. The ms-LST uses the data from neighboring stimuli to construct a higher precision transformation matrix for each stimulus to reduce the differences between transformed data and real data. Secondly, to further optimize the constructed spatial filters and reference templates, we use an online adaptation scheme to learn more features of the EEG signals of the target subject through an iterative process trial-by-trial. RESULTS: ms-LST-OA performance was measured for three datasets (Benchmark Dataset, BETA Dataset, and UCSD Dataset). Using few calibration data, the ITR of ms-LST-OA achieved 210.01±10.10 bits/min, 172.31±7.26 bits/min, and 139.04±14.90 bits/min for all three datasets, respectively. CONCLUSION: Using ms-LST-OA can reduce calibration effort for SSVEP-based BCIs.

Predictors of cancer-specific survival and overall survival among patients aged ≥60 years with lung adenocarcinoma using the SEER database.

OBJECTIVE: We developed a simple, rapid predictive model to evaluate the prognosis of older patients with lung adenocarcinoma. METHODS: Demographic characteristics and clinical information of patients with lung adenocarcinoma aged ≥60 years were retrospectively analyzed using Surveillance, Epidemiology, and End Results (SEER) data. We built nomograms of overall survival and cancer-specific survival using Cox single-factor and multi-factor regression. We used the C-index, calibration curve, receiver operating characteristic (ROC) curves, and decision curve analysis (DCA) to evaluate performance of the nomograms. RESULTS: We included 14,117 patients, divided into a training set and validation set. We used the chi-square test to compare baseline data between groups and found no significant differences. We used Cox regression analysis to screen out independent prognostic factors affecting survival time and used these factors to construct the nomogram. The ROC curve, calibration curve, C-index, and DCA curve were used to verify the model. The final results showed that our predictive model had good predictive ability, and showed better predictive ability compared with tumor-node-metastasis (TNM) staging. We also achieved good results using data of our center for external verification. CONCLUSION: The present nomogram could accurately predict prognosis in older patients with lung adenocarcinoma.

Developing and validating a nomogram for penile cancer survival: A comprehensive study based on SEER and Chinese data.

OBJECTIVE: The primary aim of this study was to create a nomogram for predicting survival outcomes in penile cancer patients, utilizing data from the Surveillance, Epidemiology, and End Results (SEER) and a Chinese organization. METHODS: Our study involved a cohort of 5744 patients diagnosed with penile cancer from the SEER database, spanning from 2004 to 2019. In addition, 103 patients with penile cancer from Sun Yat-sen Memorial Hospital of Sun Yat-sen University were included during the same period. Based on the results of regression analysis, a nomogram is constructed and validated internally and externally. The predictive performance of the model was evaluated by concordance index (c-index), area under the curve, decision curve analysis, and calibration curve, in internal and external datasets. Finally, the prediction efficiency is compared with the TNM staging model. RESULTS: A total of 3154 penile patients were randomly divided into the training group and the internal validation group at a ratio of 2:1. Nine independent risk factors were identified, including age, race, marital status, tumor grade, histology, TNM stage, and the surgical approach. Based on these factors, a nomogram was constructed to predict OS. The nomogram demonstrated relatively better consistency, predictive accuracy, and clinical relevance, with a c-index over 0.73 (in the training cohort, the validation cohort, and externally validation cohort.) These evaluation indexes are far better than the TNM staging system. CONCLUSION: Penile cancer, often overlooked in research, has lacked detailed investigative focus and guidelines. This study stands as the first to validate penile cancer prognosis using extensive data from the SEER database, supplemented by data from our own institution. Our findings equip surgeons with an essential tool to predict the prognosis of penile cancer better suited than TNM, thereby enhancing clinical decision-making processes.