Integrating Mechanistic and Toxicokinetic Information in Predictive Models of Cholestasis.

Drug development involves the thorough assessment of the candidate's safety and efficacy. In silico toxicology (IST) methods can contribute to the assessment, complementing in vitro and in vivo experimental methods, since they have many advantages in terms of cost and time. Also, they are less demanding concerning the requirements of product and experimental animals. One of these methods, Quantitative Structure-Activity Relationships (QSAR), has been proven successful in predicting simple toxicity end points but has more difficulties in predicting end points involving more complex phenomena. We hypothesize that QSAR models can produce better predictions of these end points by combining multiple QSAR models describing simpler biological phenomena and incorporating pharmacokinetic (PK) information, using quantitative in vitro to in vivo extrapolation (QIVIVE) models. In this study, we applied our methodology to the prediction of cholestasis and compared it with direct QSAR models. Our results show a clear increase in sensitivity. The predictive quality of the models was further assessed to mimic realistic conditions where the query compounds show low similarity with the training series. Again, our methodology shows clear advantages over direct QSAR models in these situations. We conclude that the proposed methodology could improve existing methodologies and could be suitable for being applied to other toxicity end points.

Automated prostate multi-regional segmentation in magnetic resonance using fully convolutional neural networks.

OBJECTIVE: Automatic MR imaging segmentation of the prostate provides relevant clinical benefits for prostate cancer evaluation such as calculation of automated PSA density and other critical imaging biomarkers. Further, automated T2-weighted image segmentation of central-transition zone (CZ-TZ), peripheral zone (PZ), and seminal vesicle (SV) can help to evaluate clinically significant cancer following the PI-RADS v2.1 guidelines. Therefore, the main objective of this work was to develop a robust and reproducible CNN-based automatic prostate multi-regional segmentation model using an intercontinental cohort of prostate MRI. METHODS: A heterogeneous database of 243 T2-weighted prostate studies from 7 countries and 10 machines of 3 different vendors, with the CZ-TZ, PZ, and SV regions manually delineated by two experienced radiologists (ground truth), was used to train (n = 123) and test (n = 120) a U-Net-based model with deep supervision using a cyclical learning rate. The performance of the model was evaluated by means of dice similarity coefficient (DSC), among others. Segmentation results with a DSC above 0.7 were considered accurate. RESULTS: The proposed method obtained a DSC of 0.88 ± 0.01, 0.85 ± 0.02, 0.72 ± 0.02, and 0.72 ± 0.02 for the prostate gland, CZ-TZ, PZ, and SV respectively in the 120 studies of the test set when comparing the predicted segmentations with the ground truth. No statistically significant differences were found in the results obtained between manufacturers or continents. CONCLUSION: Prostate multi-regional T2-weighted MR images automatic segmentation can be accurately achieved by U-Net like CNN, generalizable in a highly variable clinical environment with different equipment, acquisition configurations, and population. KEY POINTS: • Deep learning techniques allows the accurate segmentation of the prostate in three different regions on MR T2w images. • Multi-centric database proved the generalization of the CNN model on different institutions across different continents. • CNN models can be used to aid on the diagnosis and follow-up of patients with prostate cancer.

Applicability of probabilistic graphical models for early detection of SARS-CoV-2 reactive antibodies after SARS-CoV-2 vaccination in hematological patients.

Prior studies of antibody response after full SARS-CoV-2 vaccination in hematological patients have confirmed lower antibody levels compared to the general population. Serological response in hematological patients varies widely according to the disease type and its status, and the treatment given and its timing with respect to vaccination. Through probabilistic machine learning graphical models, we estimated the conditional probabilities of having detectable anti-SARS-CoV-2 antibodies at 3-6 weeks after SARS-CoV-2 vaccination in a large cohort of patients with several hematological diseases (n= 1166). Most patients received mRNA-based vaccines (97%), mainly Moderna® mRNA-1273 (74%) followed by Pfizer-BioNTech® BNT162b2 (23%). The overall antibody detection rate at 3 to 6 weeks after full vaccination for the entire cohort was 79%. Variables such as type of disease, timing of anti-CD20 monoclonal antibody therapy, age, corticosteroids therapy, vaccine type, disease status, or prior infection with SARS-CoV-2 are among the most relevant conditions influencing SARS-CoV-2-IgG-reactive antibody detection. A lower probability of having detectable antibodies was observed in patients with B-cell non-Hodgkin's lymphoma treated with anti-CD20 monoclonal antibodies within 6 months before vaccination (29.32%), whereas the highest probability was observed in younger patients with chronic myeloproliferative neoplasms (99.53%). The Moderna® mRNA-1273 compound provided higher probabilities of antibody detection in all scenarios. This study depicts conditional probabilities of having detectable antibodies in the whole cohort and in specific scenarios such as B cell NHL, CLL, MM, and cMPN that may impact humoral responses. These results could be useful to focus on additional preventive and/or monitoring interventions in these highly immunosuppressed hematological patients.

Exploring the Heterogeneity and Trajectories of Positive Functioning Variables, Emotional Distress, and Post-traumatic Growth During Strict Confinement Due to COVID-19.

COVID-19 pandemic-related confinement may be a fruitful opportunity to use individual resources to deal with it or experience psychological functioning changes. This study aimed to analyze the evolution of different psychological variables during the first coronavirus wave to identify the different psychological response clusters, as well as to keep a follow-up on the changes among these clusters. The sample included 459 Spanish residents (77.8% female, Mage = 35.21 years, SDage = 13.00). Participants completed several online self-reported questionnaires to assess positive functioning variables (MLQ, Steger et al. in J Loss Trauma 13(6):511-527, 2006. 10.1080/15325020802173660; GQ-6, McCullough et al. in J Person Soc Psychol 82:112-127, 2002. 10.1037/0022-3514.82.1.112; CD-RISC, Campbell-Sills and Stein in J Traum Stress 20(6):1019-1028, 2007. 10.1002/jts.20271; CLS-H, Chiesi et al. in BMC Psychol 8(1):1-9, 2020. 10.1186/s40359-020-0386-9; SWLS; Diener et al. in J Person Assess, 49(1), 71-75, 1985), emotional distress (PHQ-2, Kroenke et al. in Med Care 41(11):1284-1292, 2003. 10.1097/01.MLR.0000093487.78664.3C; GAD-2, Kroenke et al. in Ann Internal Med 146(5):317-325, 2007. 10.7326/0003-4819-146-5-200703060-00004; PANAS, Watson et al. in J Person Soc Psychol 47:1063-1070, 1988; Perceived Stress, ad hoc), and post-traumatic growth (PTGI-SF; Cann et al. in Anxiety Stress Coping 23(2):127-137, 2010. 10.1080/10615800903094273), four times throughout the 3 months of the confinement. Linear mixed models showed that the scores on positive functioning variables worsened from the beginning of the confinement, while emotional distress and personal strength improved by the end of the state of alarm. Clustering analyses revealed four different patterns of psychological response: "Survival", "Resurgent", "Resilient", and "Thriving" individuals. Four different profiles were identified during mandatory confinement and most participants remained in the same cluster. The "Resilient" cluster gathered the largest number of individuals (30-37%). We conclude that both the heterogeneity of psychological profiles and analysis of positive functioning variables, emotional distress, and post-traumatic growth must be considered to better understand the response to prolonged adverse situations. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s10902-021-00469-z.

Lead Reconstruction Using Artificial Neural Networks for Ambulatory ECG Acquisition.

One of the most powerful techniques to diagnose cardiovascular diseases is to analyze the electrocardiogram (ECG). To increase diagnostic sensitivity, the ECG might need to be acquired using an ambulatory system, as symptoms may occur during a patient's daily life. In this paper, we propose using an ambulatory ECG (aECG) recording device with a low number of leads and then estimating the views that would have been obtained with a standard ECG location, reconstructing the complete Standard 12-Lead System, the most widely used system for diagnosis by cardiologists. Four approaches have been explored, including Linear Regression with ECG segmentation and Artificial Neural Networks (ANN). The best reconstruction algorithm is based on ANN, which reconstructs the actual ECG signal with high precision, as the results bring a high accuracy (RMS Error < 13 µV and CC > 99.7%) for the set of patients analyzed in this paper. This study supports the hypothesis that it is possible to reconstruct the Standard 12-Lead System using an aECG recording device with less leads.

Assessment of Kinect V2 for elbow range of motion estimation in people with haemophilia using an angle correction model.

INTRODUCTION: The joint range of motion (ROM) is an important clinical parameter used to assess the loss of functionality resulting from joint bleedings in people with haemophilia. These episodes require a close follow-up and, to decrease patients' hospital dependence, telemedicine tools are needed. Therefore, this study is aimed to analyse the validity of the Microsoft Kinect V2 sensor with corrected angle measurement to be used in the monitoring of elbow ROM in people with haemophilia. METHODS: A convenience sample of 10 healthy controls (CG) and 10 patients with haemophilia with elbow arthropathy (HG) participated in this study. Full ROM of elbow joints was measured in the frontal view with a 10-degree sweep using: (a) a clinical goniometer; (b) the Kinect V2; (c) the Kinect V2 with angle correction; and (d) using a photograph. Bland-Altman graphs (mean and 95% Limits of Agreement [LOA]) and Wilcoxon test were used to determine differences between measurements and groups. RESULTS: The angle-corrected Kinect V2 measurement removed the skew in the original data, reducing the average errors from 7.9° (LoA = -10.3°; 26.0°; CG) and 9.5° (LoA = -7.9°; 26.9°; HG) to -0.1° (LoA = -8.1°; 7.9°; CG) and -0.7° (LoA = -10.7°; 9.3°; HG). CONCLUSIONS: These error levels allow the use of Kinect V2 in the clinical practice. Kinect V2 with angle correction can complement the classical goniometry allowing an efficient and touchless measurement of ROM.

Vibration Monitoring of the Mechanical Harvesting of Citrus to Improve Fruit Detachment Efficiency.

The introduction of a mechanical harvesting process for oranges can contribute to enhancing farm profitability and reducing labour dependency. The objective of this work is to determine the spread of the vibration in citrus tree canopies to establish recommendations to reach high values of fruit detachment efficiency and eliminate the need for subsequent hand-harvesting processes. Field tests were carried out with a lateral tractor-drawn canopy shaker on four commercial plots of sweet oranges. Canopy vibration during the harvesting process was measured with a set of triaxial accelerometer sensors with a datalogger placed on 90 bearing branches. Monitoring of the vibration process, fruit production, and branch properties were analysed. The improvement of fruit detachment efficiency was possible if both the hedge tree and the machinery were mutually adjusted. The hedge should be trained to facilitate access of the rods and to encourage external fructification since the internal canopy branches showed 43% of the acceleration vibration level of the external branches. The machine should be adjusted to vibrate the branches at a vibration time of at least 5.8 s, after the interaction of the rod with the branch, together with a root mean square acceleration value of 23.9 m/s2 to a complete process of fruit detachment.

HemoKinect: A Microsoft Kinect V2 Based Exergaming Software to Supervise Physical Exercise of Patients with Hemophilia.

Patients with hemophilia need to strictly follow exercise routines to minimize their risk of suffering bleeding in joints, known as hemarthrosis. This paper introduces and validates a new exergaming software tool called HemoKinect that intends to keep track of exercises using Microsoft Kinect V2's body tracking capabilities. The software has been developed in C++ and MATLAB. The Kinect SDK V2.0 libraries have been used to obtain 3D joint positions from the Kinect color and depth sensors. Performing angle calculations and center-of-mass (COM) estimations using these joint positions, HemoKinect can evaluate the following exercises: elbow flexion/extension, knee flexion/extension (squat), step climb (ankle exercise) and multi-directional balance based on COM. The software generates reports and progress graphs and is able to directly send the results to the physician via email. Exercises have been validated with 10 controls and eight patients. HemoKinect successfully registered elbow and knee exercises, while displaying real-time joint angle measurements. Additionally, steps were successfully counted in up to 78% of the cases. Regarding balance, differences were found in the scores according to the difficulty level and direction. HemoKinect supposes a significant leap forward in terms of exergaming applicability to rehabilitation of patients with hemophilia, allowing remote supervision.

Support vector machine classification of brain metastasis and radiation necrosis based on texture analysis in MRI.

PURPOSE: To develop a classification model using texture features and support vector machine in contrast-enhanced T1-weighted images to differentiate between brain metastasis and radiation necrosis. METHODS: Texture features were extracted from 115 lesions: 32 of them previously diagnosed as radiation necrosis, 23 as radiation-treated metastasis and 60 untreated metastases; including a total of 179 features derived from six texture analysis methods. A feature selection technique based on support vector machine was used to obtain a subset of features that provide optimal performance. RESULTS: The highest classification accuracy evaluated over test sets was achieved with a subset of ten features when the untreated metastases were not considered; and with a subset of seven features when the classifier was trained with untreated metastases and tested on treated ones. Receiver operating characteristic curves provided area-under-the-curve (mean ± standard deviation) of 0.94 ± 0.07 in the first case, and 0.93 ± 0.02 in the second. CONCLUSION: High classification accuracy (AUC > 0.9) was obtained using texture features and a support vector machine classifier in an approach based on conventional MRI to differentiate between brain metastasis and radiation necrosis.

A machine learning approach to identify groups of patients with hematological malignant disorders.

BACKGROUND AND OBJECTIVE: Vaccination against SARS-CoV-2 in immunocompromised patients with hematologic malignancies (HM) is crucial to reduce the severity of COVID-19. Despite vaccination efforts, over a third of HM patients remain unresponsive, increasing their risk of severe breakthrough infections. This study aims to leverage machine learning's adaptability to COVID-19 dynamics, efficiently selecting patient-specific features to enhance predictions and improve healthcare strategies. Highlighting the complex COVID-hematology connection, the focus is on interpretable machine learning to provide valuable insights to clinicians and biologists. METHODS: The study evaluated a dataset with 1166 patients with hematological diseases. The output was the achievement or non-achievement of a serological response after full COVID-19 vaccination. Various machine learning methods were applied, with the best model selected based on metrics such as the Area Under the Curve (AUC), Sensitivity, Specificity, and Matthew Correlation Coefficient (MCC). Individual SHAP values were obtained for the best model, and Principal Component Analysis (PCA) was applied to these values. The patient profiles were then analyzed within identified clusters. RESULTS: Support vector machine (SVM) emerged as the best-performing model. PCA applied to SVM-derived SHAP values resulted in four perfectly separated clusters. These clusters are characterized by the proportion of patients that generate antibodies (PPGA). Cluster 1, with the second-highest PPGA (69.91%), included patients with aggressive diseases and factors contributing to increased immunodeficiency. Cluster 2 had the lowest PPGA (33.3%), but the small sample size limited conclusive findings. Cluster 3, representing the majority of the population, exhibited a high rate of antibody generation (84.39%) and a better prognosis compared to cluster 1. Cluster 4, with a PPGA of 66.33%, included patients with B-cell non-Hodgkin's lymphoma on corticosteroid therapy. CONCLUSIONS: The methodology successfully identified four separate patient clusters using Machine Learning and Explainable AI (XAI). We then analyzed each cluster based on the percentage of HM patients who generated antibodies after COVID-19 vaccination. The study suggests the methodology's potential applicability to other diseases, highlighting the importance of interpretable ML in healthcare research and decision-making.

Usage of model combination in computational toxicology.

New Approach Methodologies (NAMs) have ushered in a new era in the field of toxicology, aiming to replace animal testing. However, despite these advancements, they are not exempt from the inherent complexities associated with the study's endpoint. In this review, we have identified three major groups of complexities: mechanistic, chemical space, and methodological. The mechanistic complexity arises from interconnected biological processes within a network that are challenging to model in a single step. In the second group, chemical space complexity exhibits significant dissimilarity between compounds in the training and test series. The third group encompasses algorithmic and molecular descriptor limitations and typical class imbalance problems. To address these complexities, this work provides a guide to the usage of a combination of predictive Quantitative Structure-Activity Relationship (QSAR) models, known as metamodels. This combination of low-level models (LLMs) enables a more precise approach to the problem by focusing on different sub-mechanisms or sub-processes. For mechanistic complexity, multiple Molecular Initiating Events (MIEs) or levels of information are combined to form a mechanistic-based metamodel. Regarding the complexity arising from chemical space, two types of approaches were reviewed to construct a fragment-based chemical space metamodel: those with and without structure sharing. Metamodels with structure sharing utilize unsupervised strategies to identify data patterns and build low-level models for each cluster, which are then combined. For situations without structure sharing due to pharmaceutical industry intellectual property, the use of prediction sharing, and federated learning approaches have been reviewed. Lastly, to tackle methodological complexity, various algorithms are combined to overcome their limitations, diverse descriptors are employed to enhance problem definition and balanced dataset combinations are used to address class imbalance issues (methodological-based metamodels). Remarkably, metamodels consistently outperformed classical QSAR models across all cases, highlighting the importance of alternatives to classical QSAR models when faced with such complexities.

Influence of genetic factors in elbow tendon pathology: a case-control study.

Elbow tendinopathy is a common pathology of the upper extremity that impacts both athletes and workers. Some research has examined the genetic component as a risk factor for tendinopathy, mainly in the lower limbs. A case-control study was designed to test for a relationship between certain collagen gene single nucleotide polymorphisms (SNPs) and elbow tendon pathology. A sample of 137 young adult athletes whose sports participation involves loading of the upper limb were examined for the presence of structural abnormalities indicative of pathology in the tendons of the lateral and medial elbow using ultrasound imaging and genotyped for the following SNPs: COL5A1 rs12722, COL11A1 rs3753841, COL11A1 rs1676486, and COL11A2 rs1799907. Anthropometric measurements and data on participants' elbow pain and dysfunction were collected using the Disabilities of the Arm, Shoulder and Hand and the Mayo Clinic Performance Index for the Elbow questionnaires. Results showed that participants in the structural abnormality group had significantly higher scores in pain and dysfunction. A significant relationship between COL11A1 rs3753841 genotype and elbow tendon pathology was found (p = 0.024), with the CT variant associated with increased risk of pathology.

Machine Learning for Mortality Analysis in Patients with COVID-19.

This paper analyzes a sample of patients hospitalized with COVID-19 in the region of Madrid (Spain). Survival analysis, logistic regression, and machine learning techniques (both supervised and unsupervised) are applied to carry out the analysis where the endpoint variable is the reason for hospital discharge (home or deceased). The different methods applied show the importance of variables such as age, O2 saturation at Emergency Rooms (ER), and whether the patient comes from a nursing home. In addition, biclustering is used to globally analyze the patient-drug dataset, extracting segments of patients. We highlight the validity of the classifiers developed to predict the mortality, reaching an appreciable accuracy. Finally, interpretable decision rules for estimating the risk of mortality of patients can be obtained from the decision tree, which can be crucial in the prioritization of medical care and resources.

Physical Activity Monitoring and Acceptance of a Commercial Activity Tracker in Adult Patients with Haemophilia.

Physical activity (PA) is highly beneficial for people with haemophilia (PWH), however, studies that objectively monitor the PA in this population are scarce. This study aimed to monitor the daily PA and analyse its evolution over time in a cohort of PWH using a commercial activity tracker. In addition, this work analyses the relationship between PA levels, demographics, and joint health status, as well as the acceptance and adherence to the activity tracker. Twenty-six PWH were asked to wear a Fitbit Charge HR for 13 weeks. According to the steps/day in the first week, data were divided into two groups: Active Group (AG; ≥10,000 steps/day) and Non-Active Group (NAG; <10,000 steps/day). Correlations between PA and patient characteristics were studied using the Pearson coefficient. Participants' user experience was analysed with a questionnaire. The 10,000 steps/day was reached by 57.7% of participants, with 12,603 (1525) and 7495 (1626) being the mean steps/day of the AG and NAG, respectively. In general, no significant variations (p > 0.05) in PA levels or adherence to wristband were produced. Only the correlation between very active minutes and arthropathy was significant (r = -0.40, p = 0.045). Results of the questionnaire showed a high level of satisfaction. In summary, PWH are able to comply with the PA recommendations, and the Fitbit wristband is a valid tool for a continuous and long-term monitoring of PA. However, by itself, the use of a wristband is not enough motivation to increase PA levels.

Effect of isolation measures on the incidence and prevalence of hepatitis C virus infection in hemodialysis.

BACKGROUND: Nosocomial transmission of hepatitis C virus (HCV) in hemodialysis (HD) units is well established. In units with a high prevalence of HCV infection, the implementation of universal precautionary measures may not suffice in order to decrease the incidence and prevalence of HCV. In this setting strict isolation practices can be useful in order to achieve this goal. METHODS: The incidence and prevalence of HCV infection amongst all HD and peritoneal dialysis (PD) patients from the province of Albacete, Spain, have been studied from 1992 to 2003.Through the 1993-1995 period chronic HD patients were treated either in a room exclusively for HCV- patients or in a room shared by HCV+ and HCV- patients. Complete separation of HCV+ and HCV- patients was implemented in 1995. Acute patients have been separated since 1992. The implementation of universal precautions was applied throughout the period. RESULTS: There has not been a single seroconversion in the rooms where only HCV- patients were dialyzed during the 11 years of follow-up. There were two seroconversions in the rooms shared for 3 years by both HCV+ and HCV- patients. In 1995 the prevalence of HCV+ cases in HD and PD was 21.6 and 23.2%, respectively. Since then it has decreased steadily and in parallel for both therapies, and the current prevalence is 6.8% in HD and 5.7% in PD. CONCLUSIONS: In HD units with a high prevalence of HCV+ patients, strict isolation in combination with implementation of universal prevention measures can eliminate nosocomial transmission and obtain a long-term reduction in prevalence.

Upport vector machines for nonlinear kernel ARMA system identification.

Nonlinear system identification based on support vector machines (SVM) has been usually addressed by means of the standard SVM regression (SVR), which can be seen as an implicit nonlinear autoregressive and moving average (ARMA) model in some reproducing kernel Hilbert space (RKHS). The proposal of this letter is twofold. First, the explicit consideration of an ARMA model in an RKHS (SVM-ARMA2K) is proposed. We show that stating the ARMA equations in an RKHS leads to solving the regularized normal equations in that RKHS, in terms of the autocorrelation and cross correlation of the (nonlinearly) transformed input and output discrete time processes. Second, a general class of SVM-based system identification nonlinear models is presented, based on the use of composite Mercer's kernels. This general class can improve model flexibility by emphasizing the input-output cross information (SVM-ARMA4K), which leads to straightforward and natural combinations of implicit and explicit ARMA models (SVR-ARMA2K and SVR-ARMA4K). Capabilities of these different SVM-based system identification schemes are illustrated with two benchmark problems.

Dosage individualization of erythropoietin using a profile-dependent support vector regression.

The external administration of recombinant human erythropoietin is the chosen treatment for those patients with secondary anemia due to chronic renal failure in periodic hemodialysis. The objective of this paper is to carry out an individualized prediction of the EPO dosage to be administered to those patients. The high cost of this medication, its side-effects and the phenomenon of potential resistance which some individuals suffer all justify the need for a model which is capable of optimizing dosage individualization. A group of 110 patients and several patient factors were used to develop the models. The support vector regressor (SVR) is benchmarked with the classical multilayer perceptron (MLP) and the Autoregressive Conditional Heteroskedasticity (ARCH) model. We introduce a priori knowledge by relaxing or tightening the epsilon-insensitive region and the penalization parameter depending on the time period of the patients' follow-up. The so-called profile-dependent SVR (PD-SVR) improves results of the standard SVR method and the MLP. We perform sensitivity analysis on the MLP and inspect the distribution of the support vectors in the input and feature spaces in order to gain knowledge about the problem.

Prediction of cyclosporine dosage in patients after kidney transplantation using neural networks.

This paper proposes the use of neural networks for individualizing the dosage of cyclosporine A (CyA) in patients who have undergone kidney transplantation. Since the dosing of CyA usually requires intensive therapeutic drug monitoring, the accurate prediction of CyA blood concentrations would decrease the monitoring frequency and, thus, improve clinical outcomes. Thirty-two patients and different factors were studied to obtain the models. Three kinds of networks (multilayer perceptron, finite impulse response (FIR) network, and Elman recurrent network) and the formation of neural-network ensembles are used in a scheme of two chained models where the blood concentration predicted by the first model constitutes an input to the dosage prediction model. This approach is designed to aid in the process of clinical decision making. The FIR network, yielding root-mean-square errors (RMSEs) of 52.80 ng/mL and mean errors (MEs) of 0.18 ng/mL in validation (10 patients) showed the best blood concentration predictions and a committee of trained networks improved the results (RMSE = 46.97 ng/mL, ME = 0.091 ng/mL). The Elman network was the selected model for dosage prediction (RMSE = 0.27 mg/Kg/d, ME = 0.07 mg/Kg/d). However, in both cases, no statistical differences on the accuracy of neural methods were found. The models' robustness is also analyzed by evaluating their performance when noise is introduced at input nodes, and it results in a helpful test for models' selection. We conclude that neural networks can be used to predict both dose and blood concentrations of cyclosporine in steady-state. This novel approach has produced accurate and validated models to be used as decision-aid tools.

Foetal ECG recovery using dynamic neural networks.

Non-invasive electrocardiography has proven to be a very interesting method for obtaining information about the foetus state and thus to assure its well-being during pregnancy. One of the main applications in this field is foetal electrocardiogram (ECG) recovery by means of automatic methods. Evident problems found in the literature are the limited number of available registers, the lack of performance indicators, and the limited use of non-linear adaptive methods. In order to circumvent these problems, we first introduce the generation of synthetic registers and discuss the influence of different kinds of noise to the modelling. Second, a method which is based on numerical (correlation coefficient) and statistical (analysis of variance, ANOVA) measures allows us to select the best recovery model. Finally, finite impulse response (FIR) and gamma neural networks are included in the adaptive noise cancellation (ANC) scheme in order to provide highly non-linear, dynamic capabilities to the recovery model. Neural networks are benchmarked with classical adaptive methods such as the least mean squares (LMS) and the normalized LMS (NLMS) algorithms in simulated and real registers and some conclusions are drawn. For synthetic registers, the most determinant factor in the identification of the models is the foetal-maternal signal-to-noise ratio (SNR). In addition, as the electromyogram contribution becomes more relevant, neural networks clearly outperform the LMS-based algorithm. From the ANOVA test, we found statistical differences between LMS-based models and neural models when complex situations (high foetal-maternal and foetal-noise SNRs) were present. These conclusions were confirmed after doing robustness tests on synthetic registers, visual inspection of the recovered signals and calculation of the recognition rates of foetal R-peaks for real situations. Finally, the best compromise between model complexity and outcomes was provided by the FIR neural network. Both the methodology for selecting a model and the introduction of advanced neural models are the main contributions of this paper.

Use of neural networks for dosage individualisation of erythropoietin in patients with secondary anemia to chronic renal failure.

The external administration of recombinant human erythropoietin is the chosen treatment for those patients with secondary anemia due to chronic renal failure undergoing periodic hemodialysis. The goal is to carry out an individualised prediction of the erythropoietin dosage to be administered. It is justified because of the high cost of this medication, its secondary effects and the phenomenon of potential resistance which some individuals suffer. One hundred and ten patients were included in this study and several factors were collected in order to develop the neural models. Since the results obtained were excellent, an easy-to-use decision-aid computer application was implemented.