Automatic Segmentation of Diffuse White Matter Abnormality on T2-weighted Brain MR Images Using Deep Learning in Very Preterm Infants

Improved Robustness for Deep Learning-based Segmentation of Multi-Center Myocardial Perfusion MRI Datasets Using Data Adaptive Uncertainty-guided Space-time Analysis

Background

Fully automatic analysis of myocardial perfusion MRI datasets enables rapid and objective reporting of stress/rest studies in patients with suspected ischemic heart disease. Developing deep learning techniques that can analyze multi-center datasets despite limited training data and variations in software (pulse sequence) and hardware (scanner vendor) is an ongoing challenge.

Methods

Datasets from 3 medical centers acquired at 3T (n = 150 subjects; 21,150 first-pass images) were included: an internal dataset (inD; n = 95) and two external datasets (exDs; n = 55) used for evaluating the robustness of the trained deep neural network (DNN) models against differences in pulse sequence (exD-1) and scanner vendor (exD-2). A subset of inD (n = 85) was used for training/validation of a pool of DNNs for segmentation, all using the same spatiotemporal U-Net architecture and hyperparameters but with different parameter initializations. We employed a space-time sliding-patch analysis approach that automatically yields a pixel-wise "uncertainty map" as a byproduct of the segmentation process. In our approach, dubbed Data Adaptive Uncertainty-Guided Space-time (DAUGS) analysis, a given test case is segmented by all members of the DNN pool and the resulting uncertainty maps are leveraged to automatically select the "best" one among the pool of solutions. For comparison, we also trained a DNN using the established approach with the same settings (hyperparameters, data augmentation, etc.).

Results

The proposed DAUGS analysis approach performed similarly to the established approach on the internal dataset (Dice score for the testing subset of inD: 0.896 ± 0.050 vs. 0.890 ± 0.049; p = n.s.) whereas it significantly outperformed on the external datasets (Dice for exD-1: 0.885 ± 0.040 vs. 0.849 ± 0.065, p < 0.005; Dice for exD-2: 0.811 ± 0.070 vs. 0.728 ± 0.149, p < 0.005). Moreover, the number of image series with "failed" segmentation (defined as having myocardial contours that include bloodpool or are noncontiguous in ≥1 segment) was significantly lower for the proposed vs. the established approach (4.3% vs. 17.1%, p < 0.0005).

Conclusions

The proposed DAUGS analysis approach has the potential to improve the robustness of deep learning methods for segmentation of multi-center stress perfusion datasets with variations in the choice of pulse sequence, site location or scanner vendor.

Structural connectivity at term equivalent age and language in preterm children at 2 years corrected

We previously reported interhemispheric structural hyperconnectivity bypassing the corpus callosum in children born extremely preterm (<28 weeks) versus term children. This increased connectivity was positively associated with language performance at 4-6 years of age in our prior work. In the present study, we aim to investigate whether this extracallosal connectivity develops in extremely preterm infants at term equivalent age by leveraging a prospective cohort study of 350 very and extremely preterm infants followed longitudinally in the Cincinnati Infant Neurodevelopment Early Prediction Study. For this secondary analysis, we included only children born extremely preterm and without significant brain injury (n = 95). We use higher-order diffusion modelling to assess the degree to which extracallosal pathways are present in extremely preterm infants and predictive of later language scores at 22-26 months corrected age. We compare results obtained from two higher-order diffusion models: generalized q-sampling imaging and constrained spherical deconvolution. Advanced MRI was obtained at term equivalent age (39-44 weeks post-menstrual age). For structural connectometry analysis, we assessed the level of correlation between white matter connectivity at the whole-brain level at term equivalent age and language scores at 2 years corrected age, controlling for post-menstrual age, sex, brain abnormality score and social risk. For our constrained spherical deconvolution analyses, we performed connectivity-based fixel enhancement, using probabilistic tractography to inform statistical testing of the hypothesis that fibre metrics at term equivalent age relate to language scores at 2 years corrected age after adjusting for covariates. Ninety-five infants were extremely preterm with no significant brain injury. Of these, 53 had complete neurodevelopmental and imaging data sets that passed quality control. In the connectometry analyses adjusted for covariates and multiple comparisons (P < 0.05), the following tracks were inversely correlated with language: bilateral cerebellar white matter and middle cerebellar peduncles, bilateral corticospinal tracks, posterior commissure and the posterior inferior fronto-occipital fasciculus. No tracks from the constrained spherical deconvolution/connectivity-based fixel enhancement analyses remained significant after correction for multiple comparisons. Our findings provide critical information about the ontogeny of structural brain networks supporting language in extremely preterm children. Greater connectivity in more posterior tracks that include the cerebellum and connections to the regions of the temporal lobes at term equivalent age appears to be disadvantageous for language development.

The past, current, and future of neonatal intensive care units with artificial intelligence: a systematic review

Machine learning and deep learning are two subsets of artificial intelligence that involve teaching computers to learn and make decisions from any sort of data. Most recent developments in artificial intelligence are coming from deep learning, which has proven revolutionary in almost all fields, from computer vision to health sciences. The effects of deep learning in medicine have changed the conventional ways of clinical application significantly. Although some sub-fields of medicine, such as pediatrics, have been relatively slow in receiving the critical benefits of deep learning, related research in pediatrics has started to accumulate to a significant level, too. Hence, in this paper, we review recently developed machine learning and deep learning-based solutions for neonatology applications. We systematically evaluate the roles of both classical machine learning and deep learning in neonatology applications, define the methodologies, including algorithmic developments, and describe the remaining challenges in the assessment of neonatal diseases by using PRISMA 2020 guidelines. To date, the primary areas of focus in neonatology regarding AI applications have included survival analysis, neuroimaging, analysis of vital parameters and biosignals, and retinopathy of prematurity diagnosis. We have categorically summarized 106 research articles from 1996 to 2022 and discussed their pros and cons, respectively. In this systematic review, we aimed to further enhance the comprehensiveness of the study. We also discuss possible directions for new AI models and the future of neonatology with the rising power of AI, suggesting roadmaps for the integration of AI into neonatal intensive care units.

White matter injury detection based on preterm infant cranial ultrasound images

Introduction

White matter injury (WMI) is now the major disease that seriously affects the quality of life of preterm infants and causes cerebral palsy of children, which also causes periventricular leuko-malacia (PVL) in severe cases. The study aimed to develop a method based on cranial ultrasound images to evaluate the risk of WMI.

Methods

This study proposed an ultrasound radiomics diagnostic system to predict the WMI risk. A multi-task deep learning model was used to segment white matter and predict the WMI risk simultaneously. In total, 158 preterm infants with 807 cranial ultrasound images were enrolled. WMI occurred in 32preterm infants (20.3%, 32/158).

Results

Ultrasound radiomics diagnostic system implemented a great result with AUC of 0.845 in the testing set. Meanwhile, multi-task deep learning model preformed a promising result both in segmentation of white matter with a Dice coefficient of 0.78 and prediction of WMI risk with AUC of 0.863 in the testing cohort.

Discussion

In this study, we presented a data-driven diagnostic system for white matter injury in preterm infants. The system combined multi-task deep learning and traditional radiomics features to achieve automatic detection of white matter regions on the one hand, and design a fusion strategy of deep learning features and manual radiomics features on the other hand to obtain stable and efficient diagnostic performance.

Characterization of MRI White Matter Signal Abnormalities in the Pediatric Population

(1) Background and Purpose: The aim of this study was to retrospectively characterize WMSAs in an unselected patient cohort at a large pediatric neuroimaging facility, in order to learn more about the spectrum of the underlying disorders encountered in everyday clinical practice. (2) Materials and Methods: Radiology reports of 5166 consecutive patients with standard brain MRI (2006-2018) were searched for predefined keywords describing WMSAs. A neuroradiology specialist enrolled patients with WMSAs following a structured approach. Imaging characteristics, etiology (autoimmune disorders, non-genetic hypoxic and ischemic insults, traumatic white matter injuries, no final diagnosis due to insufficient clinical information, "non-specific" WMSAs, infectious white matter damage, leukodystrophies, toxic white matter injuries, inborn errors of metabolism, and white matter damage caused by tumor infiltration/cancer-like disease), and age/gender distribution were evaluated. (3) Results: Overall, WMSAs were found in 3.4% of pediatric patients scanned at our and referring hospitals within the ten-year study period. The majority were found in the supratentorial region only (87%) and were non-enhancing (78% of CE-MRI). WMSAs caused by autoimmune disorders formed the largest group (23%), followed by "non-specific" WMSAs (18%), as well as non-genetic hypoxic and ischemic insults (17%). The majority were therefore acquired as opposed to inherited. Etiology-based classification of WMSAs was affected by age but not by gender. In 17% of the study population, a definite diagnosis could not be established due to insufficient clinical information (mostly external radiology consults). (4) Conclusions: An "integrated diagnosis" that combines baseline demographics, including patient age as an important factor, clinical characteristics, and additional diagnostic workup with imaging patterns can be made in the majority of cases.

Diffuse excessive high signal intensity in the preterm brain on advanced MRI represents widespread neuropathology

Preterm brains commonly exhibit elevated signal intensity in the white matter on T2-weighted MRI at term-equivalent age. This signal, known as diffuse excessive high signal intensity (DEHSI) or diffuse white matter abnormality (DWMA) when quantitatively assessed, is associated with abnormal microstructure on diffusion tensor imaging. However, postmortem data are largely lacking and difficult to obtain, and the pathological significance of DEHSI remains in question. In a cohort of 202 infants born preterm at ≤32 weeks gestational age, we leveraged two newer diffusion MRI models - Constrained Spherical Deconvolution (CSD) and neurite orientation dispersion and density index (NODDI) - to better characterize the macro and microstructural properties of DWMA and inform the ongoing debate around the clinical significance of DWMA. With increasing DWMA volume, fiber density broadly decreased throughout the white matter and fiber cross-section decreased in the major sensorimotor tracts. Neurite orientation dispersion decreased in the centrum semiovale, corona radiata, and temporal lobe. These findings provide insight into DWMA's biological underpinnings and demonstrate that it is a serious pathology.

Generalizability of Machine Learning Models: Quantitative Evaluation of Three Methodological Pitfalls

Purpose

To investigate the impact of the following three methodological pitfalls on model generalizability: (a) violation of the independence assumption, (b) model evaluation with an inappropriate performance indicator or baseline for comparison, and (c) batch effect.

Materials and Methods

The authors used retrospective CT, histopathologic analysis, and radiography datasets to develop machine learning models with and without the three methodological pitfalls to quantitatively illustrate their effect on model performance and generalizability. F1 score was used to measure performance, and differences in performance between models developed with and without errors were assessed using the Wilcoxon rank sum test when applicable.

Results

Violation of the independence assumption by applying oversampling, feature selection, and data augmentation before splitting data into training, validation, and test sets seemingly improved model F1 scores by 71.2% for predicting local recurrence and 5.0% for predicting 3-year overall survival in head and neck cancer and by 46.0% for distinguishing histopathologic patterns in lung cancer. Randomly distributing data points for a patient across datasets superficially improved the F1 score by 21.8%. High model performance metrics did not indicate high-quality lung segmentation. In the presence of a batch effect, a model built for pneumonia detection had an F1 score of 98.7% but correctly classified only 3.86% of samples from a new dataset of healthy patients.

Conclusion

Machine learning models developed with these methodological pitfalls, which are undetectable during internal evaluation, produce inaccurate predictions; thus, understanding and avoiding these pitfalls is necessary for developing generalizable models.Keywords: Random Forest, Diagnosis, Prognosis, Convolutional Neural Network (CNN), Medical Image Analysis, Generalizability, Machine Learning, Deep Learning, Model Evaluation Supplemental material is available for this article. Published under a CC BY 4.0 license.