Radiomics prognostic analysis of PET/CT images in a multicenter head and neck cancer cohort: investigating ComBat strategies, sub-volume characterization, and automatic segmentation.

PURPOSE: This study aimed to investigate the impact of several ComBat harmonization strategies, intra-tumoral sub-volume characterization, and automatic segmentations for progression-free survival (PFS) prediction through radiomics modeling for patients with head and neck cancer (HNC) in PET/CT images. METHODS: The HECKTOR MICCAI 2021 challenge set containing PET/CT images and clinical data of 325 oropharynx HNC patients was exploited. A total of 346 IBSI-compliant radiomic features were extracted for each patient's primary tumor volume defined by the reference manual contours. Modeling relied on least absolute shrinkage Cox regression (Lasso-Cox) for feature selection (FS) and Cox proportional-hazards (CoxPH) models were built to predict PFS. Within this methodological framework, 8 different strategies for ComBat harmonization were compared, including before or after FS, in feature groups separately or all features directly, and with center or clustering-determined labels. Features extracted from tumor sub-volume clustering were also investigated for their prognostic additional value. Finally, 3 automatic segmentations (2 threshold-based and a 3D U-Net) were also compared. All results were evaluated with the concordance index (C-index). RESULTS: Radiomics features without harmonization, combined with clinical factors, led to models with C-index values of 0.69 in the testing set. The best version of ComBat harmonization, i.e., after FS, for feature groups separately and relying on clustering-determined labels, achieved a C-index of 0.71. The use of features extracted from tumor sub-volumes further improved the C-index to 0.72. Models that relied on the automatic segmentations yielded close but slightly lower prognostic performance (0.67-0.70) compared to reference contours. CONCLUSION: A standard radiomics pipeline allowed for prediction of PFS in a multicenter HNC cohort. Applying a specific strategy of ComBat harmonization improved the performance. The extraction of intra-tumoral sub-volume features and automatic segmentation could contribute to the improvement and automation of prognosis modeling, respectively.

Identification of new factors associated to walking impairment in patients with vascular-type claudication.

OBJECTIVES: Mechanisms of walking limitation in arterial claudication are incompletely elucidated. We aimed to identify new variables associated to walking limitation in patients with claudication. METHODS: We retrospectively analyzed data of 1120 patients referred for transcutaneous exercise oxygen pressure recordings (TcpO2). The outcome measurement was the absolute walking time on treadmill (3.2 km/h, 10% slope). We used both: linear regression analysis and a non-linear analysis, combining support vector machines and genetic explanatory in 800 patients with the following resting variables: age, gender, body mass index, the presence of diabetes, minimal ankle to brachial index at rest, usual walking speed over 10 m (usual-pace), number of comorbid conditions, active smoking, resting heart rate, pre-test glycaemia and hemoglobin, beta-blocker use, and exercise-derived variables: minimal value of pulse oximetry, resting chest-TcpO2, decrease in chest TcpO2 during exercise, presence of buttock ischemia defined as a decrease from rest of oxygen pressure index ≤15 mmHg. We tested the models over 320 other patients. RESULTS: Independent variables associated to walking time, by decreasing importance in the models, were: age, ankle to brachial index, usual-pace; resting TcpO2, body mass index, smoking, buttock ischemia, heart rate and beta-blockers for the linear regression analysis, and were ankle to brachial index, age, body mass index, usual-pace, decrease in chest TcpO2, smoking, buttock ischemia, glycaemia, heart rate for the non-linear analysis. Testing of models over 320 new patients gave r = 0.509 for linear and 0.575 for non-linear analysis (both p < 0.05). CONCLUSION: Buttock ischemia, heart rate and usual-pace are new variables associated to walking time.

Enhancing histopathological image classification of invasive ductal carcinoma using hybrid harmonization techniques.

This study aims to develop a robust pipeline for classifying invasive ductal carcinomas and benign tumors in histopathological images, addressing variability within and between centers. We specifically tackle the challenge of detecting atypical data and variability between common clusters within the same database. Our feature engineering-based pipeline comprises a feature extraction step, followed by multiple harmonization techniques to rectify intra- and inter-center batch effects resulting from image acquisition variability and diverse patient clinical characteristics. These harmonization steps facilitate the construction of more robust and efficient models. We assess the proposed pipeline's performance on two public breast cancer databases, BreaKHIS and IDCDB, utilizing recall, precision, and accuracy metrics. Our pipeline outperforms recent models, achieving 90-95% accuracy in classifying benign and malignant tumors. We demonstrate the advantage of harmonization for classifying patches from different databases. Our top model scored 94.7% for IDCDB and 95.2% for BreaKHis, surpassing existing feature engineering-based models (92.1% for IDCDB and 87.7% for BreaKHIS) and attaining comparable performance to deep learning models. The proposed feature-engineering-based pipeline effectively classifies malignant and benign tumors while addressing variability within and between centers through the incorporation of various harmonization techniques. Our findings reveal that harmonizing variabilities between patches from different batches directly impacts the learning and testing performance of classification models. This pipeline has the potential to enhance breast cancer diagnosis and treatment and may be applicable to other diseases.

Overview on prediction, detection, and classification of atrial fibrillation using wavelets and AI on ECG.

Atrial fibrillation (AF) is the most common supraventricular cardiac arrhythmia, resulting in high mortality rates among affected patients. AF occurs as episodes coming from irregular excitations of the ventricles that affect the functionality of the heart and can increase the risk of stroke and heart attack. Early and automatic prediction, detection, and classification of AF are important steps for effective treatment. For this reason, it is the subject of intensive research in both medicine and engineering fields. The latter research focuses on three axes: prediction, classification, and detection. Knowing that AF is often asymptomatic and that its episodes are often very short, its automatic early detection is a very complicated but clinically important task to improve AF treatment and reduce the risks for the patients. This article is a review of publications from the past decade, focusing on AF episode prediction, detection, and classification using wavelets and artificial intelligence (AI). Forty-five articles were selected of which five are about AF in general, four articles compare accuracy, recall and precision between Fourier transform (FT) and wavelets transform (WT), and thirty-six are about detection, classification, and prediction of AF with WT: 15 are based on deep learning (DL) and 21 on conventional machine learning (ML). Of the thirty-six studies, thirty were published after 2015, confirming that this particular research area is very important and has great potential for future research.

Lung and colon cancer classification using medical imaging: a feature engineering approach.

Lung and colon cancers lead to a significant portion of deaths. Their simultaneous occurrence is uncommon, however, in the absence of early diagnosis, the metastasis of cancer cells is very high between these two organs. Currently, histopathological diagnosis and appropriate treatment are the only way to improve the chances of survival and reduce cancer mortality. Using artificial intelligence in the histopathological diagnosis of colon and lung cancer can provide significant help to specialists in identifying cases of colon and lung cancers with less effort, time and cost. The objective of this study is to set up a computer-aided diagnostic system that can accurately classify five types of colon and lung tissues (two classes for colon cancer and three classes for lung cancer) by analyzing their histopathological images. Using machine learning, features engineering and image processing techniques, the six models XGBoost, SVM, RF, LDA, MLP and LightGBM were used to perform the classification of histopathological images of lung and colon cancers that were acquired from the LC25000 dataset. The main advantage of using machine learning models is that they allow a better interpretability of the classification model since they are based on feature engineering; however, deep learning models are black box networks whose working is very difficult to understand due to the complex network design. The acquired experimental results show that machine learning models give satisfactory results and are very precise in identifying classes of lung and colon cancer subtypes. The XGBoost model gave the best performance with an accuracy of 99% and a F1-score of 98.8%. The implementation and the development of this model will help healthcare specialists identify types of colon and lung cancers. The code will be available upon request.

A hierarchical modeling approach of hippocampus local circuit.

The modeling and simulation of a realistic nervous tissue are difficult because of the number of implied cell types (neuronal and glial), the topology of the networks, and the various heterogeneous molecular mechanisms. The MTIP (Mathematical Theory of Integrative Physiology) is used as a new modeling approach based on a representation in terms of functional interactions and a formalism (S-Propagator) related to n-level field theory. This work presents the passage from a theoretical description of the biological system to a computing implementation in the general case. The specific case of the hippocampus is presented, as well as how a drug allows learning and memory improvement in the local circuit of the CA1 area of the hippocampus. This in silico result is used to experimentally predict the drug effect in vitro to confirm the accuracy of MTIP.

Mathematical constraints for building learning rules in the Purkinje cell system.

We first present a method to mathematically build a learning rule for closed-loop neural networks. This rule is then applied to climbing fibers in the cerebellar cortex. Our analytical study is based on previous experimental non-analytical studies, which suggests that climbing fibers carry out an error signal to the brain. Thus, our goal is to find the class of functions for the activity propagated by climbing fibers, allowing the output of the Purkinje cell to converge towards a desired output. These functions must tend towards zero when the objective is reached. Our techniques are generalized to other network models.

On the integration of physiological mechanisms in the nervous tissue using the MTIP: synaptic plasticity depending on neurons-astrocytes-capillaries interactions.

The objective in this work is twofold: (i) to illustrate the use of the Mathematical Theory of Integrative Physiology (MTIP) [13], that is a general theory and practical method for the systematic and progressive mathematical integration of physiological mechanisms; (ii) to study a complex neurobiological system taken as an example, i.e., the synaptic plasticity depending on brain activity, on astrocytic and neuronal metabolism, and on brain hemodynamics. The functional organization of the nervous tissue is presented in the framework of the MTIP, the ultimate objective being the study of learning and memory by coupling the three networks of neurons, astrocytes and capillaries. Specifically in this paper, we study the influence of the variation of capillaries arterial oxygen on the induction of LTP/LTD by coupling validated mathematical models of AMPA, NMDA, VDCC channels, calcium current in the dendritic spine, vesicular glutamate dynamics in the presynaptic bouton derived from glycolysis and neuronal glucose, mitochondrial respiration, Ca/Na pumps, glycolysis, and calcium dynamics in the astrocytes, hemodynamics of the capillaries. The integration of all these models is discussed by claiming the advantages of using a common framework and a specific dedicated computing system, PhysioMatica.

On the mathematical integration of the nervous tissue based on the S-propagator formalism: II. Numerical simulations for molecular-dependent activity.

In a previous article (G. A. Chauvet, 2002), presenting a theoretical approach for integrating physiological functions in nervous tissue, we showed that a specific hierarchical representation, incorporating the novel concepts of non-symmetry and non-locality, and an appropriate formalism (the S-propagator formalism) could provide a good description of a living system in general, and the nervous system in particular. We now show that, in the framework of this theory, in spite of the complexity inherent to nervous tissue and the great number of elementary mechanisms involved, the numerical resolution of the global non-local system allows us to envisage simulations that would otherwise be impossible to realize. Here, the study is limited to one physiological function, i.e., the spatiotemporal variation of membrane potential in neuronal tissue. We demonstrate that the role of the kinetic constants at the molecular level is in agreement with the observed activity of the neuronal network. The method also reveals the critical role of the maximum density of synapses along the dendritic tree in the behavior of the network. This illustrates the great advantage of the theoretical approach in studying separately any other complementary coupled function without having to modify the computational methods used here. The application of this method to the spatiotemporal variation of synaptic efficacy, which is the basis of the learning and memory function, will be treated in a forthcoming paper.

Evaluation of automatic feature detection algorithms in EEG: application to interburst intervals.

In this paper, we present a new method to compare and improve algorithms for feature detection in neonatal EEG. The method is based on the algorithm׳s ability to compute accurate statistics to predict the results of EEG visual analysis. This method is implemented inside a Java software called EEGDiag, as part of an e-health Web portal dedicated to neonatal EEG. EEGDiag encapsulates a component-based implementation of the detection algorithms called analyzers. Each analyzer is defined by a list of modules executed sequentially. As the libraries of modules are intended to be enriched by its users, we developed a process to evaluate the performance of new modules and analyzers using a database of expertized and categorized EEGs. The evaluation is based on the Davies-Bouldin index (DBI) which measures the quality of cluster separation, so that it will ease the building of classifiers on risk categories. For the first application we tested this method on the detection of interburst intervals (IBI) using a database of 394 EEG acquired on premature newborns. We have defined a class of IBI detectors based on a threshold of the standard deviation on contiguous short time windows, inspired by previous work. Then we determine which detector and what threshold values are the best regarding DBI, as well as the robustness of this choice. This method allows us to make counter-intuitive choices, such as removing the 50 Hz filter (power supply) to save time.