Autonomous fetal morphology scan: deep learning + clustering merger - the second pair of eyes behind the doctor.

The main cause of fetal death, of infant morbidity or mortality during childhood years is attributed to congenital anomalies. They can be detected through a fetal morphology scan. An experienced sonographer (with more than 2000 performed scans) has the detection rate of congenital anomalies around 52%. The rates go down in the case of a junior sonographer, that has the detection rate of 32.5%. One viable solution to improve these performances is to use Artificial Intelligence. The first step in a fetal morphology scan is represented by the differentiation process between the view planes of the fetus, followed by a segmentation of the internal organs in each view plane. This study presents an Artificial Intelligence empowered decision support system that can label anatomical organs using a merger between deep learning and clustering techniques, followed by an organ segmentation with YOLO8. Our framework was tested on a fetal morphology image dataset that regards the fetal abdomen. The experimental results show that the system can correctly label the view plane and the corresponding organs on real-time ultrasound movies.Trial registrationThe study is registered under the name "Pattern recognition and Anomaly Detection in fetal morphology using Deep Learning and Statistical Learning (PARADISE)", project number 101PCE/2022, project code PN-III-P4-PCE-2021-0057. Trial registration: ClinicalTrials.gov, unique identifying number NCT05738954, date of registration 02.11.2023.

Pattern Recognition and Anomaly Detection in fetal morphology using Deep Learning and Statistical learning (PARADISE): protocol for the development of an intelligent decision support system using fetal morphology ultrasound scan to detect fetal congenital anomaly detection.

INTRODUCTION: Congenital anomalies are the most encountered cause of fetal death, infant mortality and morbidity. 7.9 million infants are born with congenital anomalies yearly. Early detection of congenital anomalies facilitates life-saving treatments and stops the progression of disabilities. Congenital anomalies can be diagnosed prenatally through morphology scans. A correct interpretation of the morphology scan allows a detailed discussion with the parents regarding the prognosis. The central feature of this project is the development of a specialised intelligent system that uses two-dimensional ultrasound movies obtained during the standard second trimester morphology scan to identify congenital anomalies in fetuses. METHODS AND ANALYSIS: The project focuses on three pillars: committee of deep learning and statistical learning algorithms, statistical analysis, and operational research through learning curves. The cross-sectional study is divided into a training phase where the system learns to detect congenital anomalies using fetal morphology ultrasound scan, and then it is tested on previously unseen scans. In the training phase, the intelligent system will learn to answer the following specific objectives: (a) the system will learn to guide the sonographer's probe for better acquisition; (b) the fetal planes will be automatically detected, measured and stored and (c) unusual findings will be signalled. During the testing phase, the system will automatically perform the above tasks on previously unseen videos.Pregnant patients in their second trimester admitted for their routine scan will be consecutively included in a 32-month study (4 May 2022-31 December 2024). The number of patients is 4000, enrolled by 10 doctors/sonographers. We will develop an intelligent system that uses multiple artificial intelligence algorithms that interact between themselves, in bulk or individual. For each anatomical part, there will be an algorithm in charge of detecting it, followed by another algorithm that will detect whether anomalies are present or not. The sonographers will validate the findings at each intermediate step. ETHICS AND DISSEMINATION: All protocols and the informed consent form comply with the Health Ministry and professional society ethics guidelines. The University of Craiova Ethics Committee has approved this study protocol as well as the Romanian Ministry of Research Innovation and Digitization that funded this research. The study will be implemented and reported in line with the STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) statement. TRIAL REGISTRATION NUMBER: The study is registered under the name 'Pattern recognition and Anomaly Detection in fetal morphology using Deep Learning and Statistical Learning', project number 101PCE/2022, project code PN-III-P4-PCE-2021-0057. TRIAL REGISTRATION: ClinicalTrials.gov, unique identifying number NCT05738954, date of registration: 2 November 2023.

Deep learning and Gaussian Mixture Modelling clustering mix. A new approach for fetal morphology view plane differentiation.

The last three years have been a game changer in the way medicine is practiced. The COVID-19 pandemic changed the obstetrics and gynecology scenery. Pregnancy complications, and even death, are preventable due to maternal-fetal monitoring. A fast and accurate diagnosis can be established by a doctor + Artificial Intelligence combo. The aim of this paper is to propose a framework designed as a merger between Deep learning algorithms and Gaussian Mixture Modelling clustering applied in differentiating between the view planes of a second trimester fetal morphology scan. The deep learning methods chosen for this approach were ResNet50, DenseNet121, InceptionV3, EfficientNetV2S, MobileNetV3Large, and Xception. The framework establishes a hierarchy of the component networks using a statistical fitness function and the Gaussian Mixture Modelling clustering method, followed by a synergetic weighted vote of the algorithms that gives the final decision. We have tested the framework on two second trimester morphology scan datasets. A thorough statistical benchmarking process has been provided to validate our results. The experimental results showed that the synergetic vote of the framework outperforms the vote of each stand-alone deep learning network, hard voting, soft voting, and bagging strategy.

Doctor/Data Scientist/Artificial Intelligence Communication Model. Case Study.

The last two years have taught us that we need to change the way we practice medicine. Due to the COVID-19 pandemic, obstetrics and gynecology setting has changed enormously. Monitoring pregnant women prevents deaths and complications. Doctors and computer data scientists must learn to communicate and work together to improve patients' health. In this paper we present a good practice example of a competitive/collaborative communication model for doctors, computer scientists and artificial intelligence systems, for signaling fetal congenital anomalies in the second trimester morphology scan.

Prediction of labor outcome pilot study: evaluation of primiparous women at term.

BACKGROUND: Emergency operative delivery is associated with high fetal and maternal morbidity and mortality. It is of high importance to find means to predict the delivery mode before the onset of labor. OBJECTIVE: This study aimed to investigate the potential of combined sonographic and clinical determination to predict the mode of delivery at term. STUDY DESIGN: An observational prospective cohort study was deployed in a tertiary maternity hospital (Emergency County Hospital Craiova). Unselected low-risk primiparous pregnant women were evaluated weekly at term for ultrasound determinations (estimated fetal weight, head descent parameters, occiput posterior, cervical length), Bishop score, and maternal characteristics (age, height, weight). A thorough statistical analysis determined which variables were significantly correlated with the delivery mode. RESULTS: Data from 276 term primiparous women were analyzed. Head descent parameters were strongly and significantly correlated with each other, but only progression distance was correlated with the delivery mode (gestational weeks 37, 38, 41, and the week before delivery). In the week before delivery, measurements of head-to-perineum distance and angle of progression reached almost significant P levels of.055 and.07, respectively. The following variables were significantly correlated with the delivery mode: body mass index in all term evaluations; progression distance for weeks 37 and 38; maternal age for week 39; Bishop score, estimated fetal weight, and occiput posterior for week 40; and body mass index, estimated fetal weight, and progression distance for the week before delivery. We also provided logistic regression equations for each week with correct delivery mode prediction, except for week 38. Cutoff values were established for each significant parameter per week. The cutoff values must be read in conjunction with the area under the curve, which ranged from 0.55 to 0.73, depending on the variable. CONCLUSION: There are strong and significant correlations among the "head descent" ultrasound measurements at term. Body mass index is predictive of labor outcomes throughout term evaluations. Progression distance and body mass index measured at 37 to 38 weeks' gestation correlate with the delivery mode and apparently can be used to forecast the delivery mode when the pregnancy reaches term. For the week before delivery, measurements of estimated fetal weight and progression distance can be used to forecast the delivery mode, perhaps as part of a policy for pregnant women with prelabor clinical signs. Larger studies with more data, particularly better-balanced data, are needed.

Learning deep neural networks' architectures using differential evolution. Case study: Medical imaging processing.

The COVID-19 pandemic has changed the way we practice medicine. Cancer patient and obstetric care landscapes have been distorted. Delaying cancer diagnosis or maternal-fetal monitoring increased the number of preventable deaths or pregnancy complications. One solution is using Artificial Intelligence to help the medical personnel establish the diagnosis in a faster and more accurate manner. Deep learning is the state-of-the-art solution for image classification. Researchers manually design the structure of fix deep learning neural networks structures and afterwards verify their performance. The goal of this paper is to propose a potential method for learning deep network architectures automatically. As the number of networks architectures increases exponentially with the number of convolutional layers in the network, we propose a differential evolution algorithm to traverse the search space. At first, we propose a way to encode the network structure as a candidate solution of fixed-length integer array, followed by the initialization of differential evolution method. A set of random individuals is generated, followed by mutation, recombination, and selection. At each generation the individuals with the poorest loss values are eliminated and replaced with more competitive individuals. The model has been tested on three cancer datasets containing MRI scans and histopathological images and two maternal-fetal screening ultrasound images. The novel proposed method has been compared and statistically benchmarked to four state-of-the-art deep learning networks: VGG16, ResNet50, Inception V3, and DenseNet169. The experimental results showed that the model is competitive to other state-of-the-art models, obtaining accuracies between 78.73% and 99.50% depending on the dataset it had been applied on.

Virtual autopsy and confirmation of normal fetal heart anatomy in the first trimester using three-dimensional (3D) reconstruction of histological sections.

OBJECTIVE: In this pilot study, we tested the feasibility of cardiac structures reconstruction from histological sections in 12-13 weeks normal fetuses. Conventional autopsy is hampered at this gestational age because of the small size of the heart anatomical structures, while alternative non-invasive methods for pathology examination of the fetus are expensive, rarely available and lack accuracy data regarding the confirmation of first trimester heart defects suspected by early prenatal ultrasound (US) scans. MATERIALS AND METHODS: Normal hearts from fetuses aged 12-13 gestational weeks (GW) were harvested for histological preparation, virtual reconstruction, and cardiac structures analysis. The normalcy of heart structures was confirmed before pregnancy termination, using a detailed US scan protocol. The fetal heart was routinely processed for formalin fixation and paraffin embedding (FFPE) and 10 µm seriate sections have been cut until finishing the specimen. All sections have been scanned and a three-dimensional (3D) reconstruction of the whole organ has been rendered, based on computer-aided manual tracing. Using the 3D navigation software, the main cardiac structures were searched for a proper and confident visualization. RESULTS: Five cases were investigated. Visualization of the normal heart cavities, including atrioventricular septum was very good in all fetuses. The entire course of right and left ventricle outflow tracts was confidently confirmed, along the branching pattern of aorta and pulmonary artery trunk. Regarding the veno-atrial connections, it was easy to identify the entrance of the inferior and superior caval veins into the right atrium, but a detailed review of the histological sections was necessary for the visualization of the left atrium venous openings. The inherent morphological deformation following heart block sectioning resulted in a lower resolution or quality of the "reconstructed" planes, but these distortions did not represent a significant impediment in any of the cases. The resources involved ordinary histology and information technology (IT) equipment. To further decrease the time involved by the protocol, many steps may be automated: cutting, coloring, and scanning. CONCLUSIONS: The results indicate that this method can be implemented to routine clinical practice. The use of 3D reconstruction of fetal heart histological sections in first trimester may serve as an important audit to confirm the normalcy of heart structures. Also, the histological and postprocessed information is retained, and this volume can be stored, reanalyzed, or sent online for a second opinion. The method involves relatively undemanding resources, i.e., hardware, software, competences, and time. The procedure could also benefit from refinements used in other imaging techniques to limit human-computer interactions, such as sections distortion.

Sonographic Evaluation of the Mechanism of Active Labor (SonoLabor Study): observational study protocol regarding the implementation of the sonopartogram.

INTRODUCTION: Over the last decades, a large body of literature has shown that intrapartum clinical digital pelvic estimations of fetal head position, station and progression in the pelvic canal are less accurate, compared with ultrasound (US) scan. Given the increasing evidence regarding the advantages of using US to evaluate the mechanism of labour, our study protocol aims to develop sonopartograms for fetal cephalic presentations. They will allow for a more objective evaluation of labour progression than the traditional labour monitoring, which could enable more rapid decisions regarding the mode of delivery. METHODS/ANALYSIS: This is a prospective observational study performed in three university hospitals, with an unselected population of women admitted in labour at term. Both clinical and US evaluations will be performed assessing fetal head position, descent and rotation. Specific US parameters regarding fetal head position, progression and rotation will be recorded to develop nomograms in a similar way that partograms were developed. The primary outcome is to develop nomograms for the longitudinal US assessment of labour in unselected nulliparous and multiparous women with fetal cephalic presentation. The secondary aims are to assess the sonopartogram differences in occiput anterior and posterior deliveries, to compare the labour trend from our research with the classic and other recent partogram models and to investigate the capability of the US labour monitoring to predict the outcome of spontaneous vaginal delivery. ETHICS AND DISSEMINATION: All protocols and the informed consent form comply with the Ministry of Health and the professional society ethics guidelines. University ethics committees approved the study protocol. The trial results will be published in peer-reviewed journals and at the conference presentations. The study will be implemented and reported in line with the Strengthening the Reporting of Observational Studies in Epidemiology statement. TRIAL REGISTRATION NUMBER: ClinicalTrials.gov Registry (NCT02326077).

Automated Gleason grading of prostate cancer using transfer learning from general-purpose deep-learning networks.

Two deep-learning algorithms designed to classify images according to the Gleason grading system that used transfer learning from two well-known general-purpose image classification networks (AlexNet and GoogleNet) were trained on Hematoxylin-Eosin histopathology stained microscopy images with prostate cancer. The dataset consisted of 439 images asymmetrically distributed in four Gleason grading groups. Mean and standard deviation accuracy for AlexNet derivate network was of 61.17±7 and for GoogleNet derivate network was of 60.9±7.4. The similar results obtained by the two networks with very different architecture, together with the normal distribution of classification error for both algorithms show that we have reached a maximum classification rate on this dataset. Taking into consideration all the constraints, we conclude that the resulted networks could assist pathologists in this field, providing first or second opinions on Gleason grading, thus presenting an objective opinion in a grading system which has showed in time a great deal of interobserver variability.

Logistic regression paradigm for training a single-hidden layer feedforward neural network. Application to gene expression datasets for cancer research.

OBJECTIVE: The speed of the diagnosis process is vital in pursuing the trial of curing cancer. During the last decade, precision medicine evolved by detecting different types of cancer through microarrays (MA) of deoxyribonucleic acid (DNA) processed by machine learning (ML) algorithms. Personalized diagnosis, followed by personalized treatment, should imply personalized hyperparameters of the ML. The goal of this paper is to propose a novel adaptive ML method that embeds knowledge into the architecture of the algorithm and also filters the features in order to reduce their number, increase computational speed, and decrease computational cost and time. MATERIALS AND METHODS: fLogSLFN is a novel two-fold theoretically effective ML that can be used in two-class decision problems that embeds the logistic regression in such a manner that the hidden nodes of a single-hidden layer feedforward neural network (SLFN) are problem dependent. A filtering module based on the significance of each attribute is embedded in order to avoid the 'curse of dimensionality' phenomenon. The proposed model has been tested on three publicly available high-dimensional cancer datasets that contain gene expressions provided by complementary DNA (cDNA) array, and DNA microarray. The proposed novel method filtered logistic SLFN (fLogSLFN) has been also compared and statistically benchmarked to four ML algorithms: extreme learning machine (ELM), radial basis function network (RBF), single-hidden layer feedforward neural network trained by the backpropagation algorithm (BPNN), logistic regression with the LASSO penalty, and the adaptive single-hidden layer feedforward network (aSLFN). MAIN FINDINGS: The experimental results showed that the fLogSLFN is competitive to the other state-of-the-art models, obtaining accuracies between 64.70% and 98.66% depending on the dataset it had been applied on. CONCLUSIONS: In contrast to other state-of-the-art ML algorithms, the fLogSLFN is capable to embed the knowledge extracted from the data into its architecture, making it problem dependent. The filtering module increases its computational speed, while decreasing computational cost and time. The statistical analysis revealed the fact that by filtering the features the performance is kept, making the algorithm more efficient.

Learning a single-hidden layer feedforward neural network using a rank correlation-based strategy with application to high dimensional gene expression and proteomic spectra datasets in cancer detection.

Methods based on microarrays (MA), mass spectrometry (MS), and machine learning (ML) algorithms have evolved rapidly in recent years, allowing for early detection of several types of cancer. A pitfall of these approaches, however, is the overfitting of data due to large number of attributes and small number of instances -- a phenomenon known as the 'curse of dimensionality'. A potentially fruitful idea to avoid this drawback is to develop algorithms that combine fast computation with a filtering module for the attributes. The goal of this paper is to propose a statistical strategy to initiate the hidden nodes of a single-hidden layer feedforward neural network (SLFN) by using both the knowledge embedded in data and a filtering mechanism for attribute relevance. In order to attest its feasibility, the proposed model has been tested on five publicly available high-dimensional datasets: breast, lung, colon, and ovarian cancer regarding gene expression and proteomic spectra provided by cDNA arrays, DNA microarray, and MS. The novel algorithm, called adaptive SLFN (aSLFN), has been compared with four major classification algorithms: traditional ELM, radial basis function network (RBF), single-hidden layer feedforward neural network trained by backpropagation algorithm (BP-SLFN), and support vector-machine (SVM). Experimental results showed that the classification performance of aSLFN is competitive with the comparison models.

Variable selection in Logistic regression model with genetic algorithm.

Variable or feature selection is one of the most important steps in model specification. Especially in the case of medical-decision making, the direct use of a medical database, without a previous analysis and preprocessing step, is often counterproductive. In this way, the variable selection represents the method of choosing the most relevant attributes from the database in order to build a robust learning models and, thus, to improve the performance of the models used in the decision process. In biomedical research, the purpose of variable selection is to select clinically important and statistically significant variables, while excluding unrelated or noise variables. A variety of methods exist for variable selection, but none of them is without limitations. For example, the stepwise approach, which is highly used, adds the best variable in each cycle generally producing an acceptable set of variables. Nevertheless, it is limited by the fact that it commonly trapped in local optima. The best subset approach can systematically search the entire covariate pattern space, but the solution pool can be extremely large with tens to hundreds of variables, which is the case in nowadays clinical data. Genetic algorithms (GA) are heuristic optimization approaches and can be used for variable selection in multivariable regression models. This tutorial paper aims to provide a step-by-step approach to the use of GA in variable selection. The R code provided in the text can be extended and adapted to other data analysis needs.

Boosting backpropagation algorithm by stimulus-sampling: Application in computer-aided medical diagnosis.

Neural networks (NNs), in general, and multi-layer perceptron (MLP), in particular, represent one of the most efficient classifiers among the machine learning (ML) algorithms. Inspired by the stimulus-sampling paradigm, it is plausible to assume that the association of stimuli with the neurons in the output layer of a MLP can increase its performance. The stimulus-sampling process is assumed memoryless (Markovian), in the sense that the choice of a particular stimulus at a certain step, conditioned by the whole prior evolution of the learning process, depends only on the network's answer at the previous step. This paper proposes a novel learning technique, by enhancing the standard backpropagation algorithm performance with the aid of a stimulus-sampling procedure applied to the output neurons. The network uses the observable behavior that varies throughout the training process by stimulating the correct answers through corresponding rewards/penalties assigned to the output neurons. The proposed model has been applied in computer-aided medical diagnosis using five real-life breast cancer, colon cancer, diabetes, thyroid, and fetal heartbeat databases. The statistical comparison to well-established ML algorithms proved beyond doubt its efficiency and robustness.

A hybrid genetic algorithm-queuing multi-compartment model for optimizing inpatient bed occupancy and associated costs.

PURPOSE: Explore how efficient intelligent decision support systems, both easily understandable and straightforwardly implemented, can help modern hospital managers to optimize both bed occupancy and utilization costs. METHODS AND MATERIALS: This paper proposes a hybrid genetic algorithm-queuing multi-compartment model for the patient flow in hospitals. A finite capacity queuing model with phase-type service distribution is combined with a compartmental model, and an associated cost model is set up. An evolutionary-based approach is used for enhancing the ability to optimize both bed management and associated costs. In addition, a "What-if analysis" shows how changing the model parameters could improve performance while controlling costs. The study uses bed-occupancy data collected at the Department of Geriatric Medicine - St. George's Hospital, London, period 1969-1984, and January 2000. RESULTS: The hybrid model revealed that a bed-occupancy exceeding 91%, implying a patient rejection rate around 1.1%, can be carried out with 159 beds plus 8 unstaffed beds. The same holding and penalty costs, but significantly different bed allocations (156 vs. 184 staffed beds, and 8 vs. 9 unstaffed beds, respectively) will result in significantly different costs (£755 vs. £1172). Moreover, once the arrival rate exceeds 7 patient/day, the costs associated to the finite capacity system become significantly smaller than those associated to an Erlang B queuing model (£134 vs. £947). CONCLUSION: Encoding the whole information provided by both the queuing system and the cost model through chromosomes, the genetic algorithm represents an efficient tool in optimizing the bed allocation and associated costs. The methodology can be extended to different medical departments with minor modifications in structure and parameterization.

Improving hospital bed occupancy and resource utilization through queuing modeling and evolutionary computation.

Scarce healthcare resources require carefully made policies ensuring optimal bed allocation, quality healthcare service, and adequate financial support. This paper proposes a complex analysis of the resource allocation in a hospital department by integrating in the same framework a queuing system, a compartmental model, and an evolutionary-based optimization. The queuing system shapes the flow of patients through the hospital, the compartmental model offers a feasible structure of the hospital department in accordance to the queuing characteristics, and the evolutionary paradigm provides the means to optimize the bed-occupancy management and the resource utilization using a genetic algorithm approach. The paper also focuses on a "What-if analysis" providing a flexible tool to explore the effects on the outcomes of the queuing system and resource utilization through systematic changes in the input parameters. The methodology was illustrated using a simulation based on real data collected from a geriatric department of a hospital from London, UK. In addition, the paper explores the possibility of adapting the methodology to different medical departments (surgery, stroke, and mental illness). Moreover, the paper also focuses on the practical use of the model from the healthcare point of view, by presenting a simulated application.

Error-correction learning for artificial neural networks using the Bayesian paradigm. Application to automated medical diagnosis.

Automated medical diagnosis models are now ubiquitous, and research for developing new ones is constantly growing. They play an important role in medical decision-making, helping physicians to provide a fast and accurate diagnosis. Due to their adaptive learning and nonlinear mapping properties, the artificial neural networks are widely used to support the human decision capabilities, avoiding variability in practice and errors based on lack of experience. Among the most common learning approaches, one can mention either the classical back-propagation algorithm based on the partial derivatives of the error function with respect to the weights, or the Bayesian learning method based on posterior probability distribution of weights, given training data. This paper proposes a novel training technique gathering together the error-correction learning, the posterior probability distribution of weights given the error function, and the Goodman-Kruskal Gamma rank correlation to assembly them in a Bayesian learning strategy. This study had two main purposes; firstly, to develop anovel learning technique based on both the Bayesian paradigm and the error back-propagation, and secondly,to assess its effectiveness. The proposed model performance is compared with those obtained by traditional machine learning algorithms using real-life breast and lung cancer, diabetes, and heart attack medical databases. Overall, the statistical comparison results indicate that thenovellearning approach outperforms the conventional techniques in almost all respects.

Evolutionary strategy to develop learning-based decision systems. Application to breast cancer and liver fibrosis stadialization.

The purpose of this paper is twofold: first, to propose an evolutionary-based method for building a decision model and, second, to assess and validate the model's performance using five different real-world medical datasets (breast cancer and liver fibrosis) by comparing it with state-of-the-art machine learning techniques. The evolutionary-inspired approach has been used to develop the learning-based decision model in the following manner: the hybridization of algorithms has been considered as "crossover", while the development of new variants which can be thought of as "mutation". An appropriate hierarchy of the component algorithms was established based on a statistically built fitness measure. A synergetic decision-making process, based on a weighted voting system, involved the collaboration between the selected algorithms in making the final decision. Well-established statistical performance measures and comparison tests have been extensively used to design and implement the model. Finally, the proposed method has been tested on five medical datasets, out of which four publicly available, and contrasted with state-of-the-art techniques, showing its efficiency in supporting the medical decision-making process.

Real-time sono-elastography in the diagnosis of diffuse liver diseases.

AIM: To analyze whether computer-enhanced dynamic analysis of elastography movies is able to better characterize and differentiate between different degrees of liver fibrosis. METHODS: The study design was prospective. A total of 132 consecutive patients with chronic liver diseases and healthy volunteers were examined by transabdominal ultrasound elastography. All examinations were done by two doctors. RESULTS: Due to the limitations of the method, we obtained high-quality elastography information in only 73.48% of the patients. The kappa-means clustering method was applied to assess the inter-observer diagnosis variability, which showed good variability values in accordance with the experience of ultrasound examination of every observer. Cohen's kappa test indicated a moderate agreement between the study observers (kappa = 0.4728). Furthermore, we compared the way the two observers clustered the patients, using the test for comparing two proportions (t value, two-sided test). There was no statistically significant difference between the two physicians, regardless of the patients' real status. CONCLUSION: Transabdominal real-time elastography is certainly a very useful method in depicting liver hardness, although it is incompletely tested in large multicenter studies.