Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

Radiology Radiology

Validation of an automated machine learning algorithm for the detection and analysis of cerebral aneurysms.

In Journal of neurosurgery ; h5-index 64.0

OBJECTIVE : Machine learning algorithms have shown groundbreaking results in neuroimaging. The authors herein evaluated the performance of a newly developed convolutional neural network (CNN) to detect and analyze intracranial aneurysms (IAs) on CTA.

METHODS : Consecutive patients with CTA studies between January 2015 and July 2021 at a single center were identified. The ground truth determination of cerebral aneurysm presence or absence was made from the neuroradiology report. The primary outcome was the performance of the CNN in detecting IAs in an external validation set, measured using area under the receiver operating characteristic curve statistics. Secondary outcomes included accuracy for location and size measurement.

RESULTS : The independent validation imaging data set consisted of 400 patients with CTA studies, median age 40 years (IQR 34 years) and 141 (35.3%) of whom were male; 193 patients (48.3%) had a diagnosis of IA on neuroradiologist evaluation. The median maximum IA diameter was 3.7 mm (IQR 2.5 mm). In the independent validation imaging data set, the CNN performed well with 93.8% sensitivity (95% CI 0.87-0.98), 94.2% specificity (95% CI 0.90-0.97), and a positive predictive value of 88.2% (95% CI 0.80-0.94) in the subgroup with an IA diameter ≥ 4 mm.

CONCLUSIONS : The described Viz.ai Aneurysm CNN performed well in identifying the presence or absence of IAs in an independent validation imaging set. Further studies are necessary to investigate the impact of the software on detection rates in a real-world setting.

Colasurdo Marco, Shalev Daphna, Robledo Ariadna, Vasandani Viren, Luna Zean Aaron, Rao Abhijit S, Garcia Roberto, Edhayan Gautam, Srinivasan Visish M, Sheth Sunil A, Donner Yoni, Bibas Orin, Limzider Nicole, Shaltoni Hashem, Kan Peter

2023-Mar-03

CT angiography, artificial intelligence, cerebral aneurysm, diagnostic technique, endovascular neurosurgery, technology, vascular disorders

General General

Multiomics and machine-learning identify novel transcriptional and mutational signatures in amyotrophic lateral sclerosis.

In Brain : a journal of neurology

Amyotrophic lateral sclerosis (ALS) is a fatal and incurable neurodegenerative disease that mainly affects the neurons of the motor system. Despite the increasing understanding of its genetic components, their biological meanings are still poorly understood. Indeed, it is still not clear to which extent the pathological features associated with ALS are commonly shared by the different genes causally linked to this disorder. To address this point, we combined multi-omics analysis covering the transcriptional, epigenetic and mutational aspects of heterogenous hiPSC-derived C9orf72-, TARDBP-, SOD1- and FUS-mutant motor neurons as well as datasets from patients' biopsies. We identified a common signature, converging toward increased stress and synaptic abnormalities, which reflects a unifying transcriptional program in ALS despite the specific profiles owing to the underlying pathogenic gene. In addition, whole genome bisulfite sequencing linked the altered gene expression observed in mutant cells to their methylation profile, highlighting deep epigenetic alterations as part of the abnormal transcriptional signatures linked to ALS. We then applied multi-layer deep machine-learning to integrate publicly-available blood and spinal cord transcriptomes and found a statistically significant correlation between their top predictor gene sets, which were significantly enriched in toll-like receptor signaling. Notably, the overrepresentation of this biological term also correlated with the transcriptional signature identified in mutant hiPSC-derived motor neurons, highlighting novel insights into ALS marker genes in a tissue-independent manner. Finally, using whole genome sequencing in combination with deep learning, we generated the first mutational signature for ALS and defined a specific genomic profile for this disease, which is significantly correlated to aging signatures, hinting at age as a major player in ALS. All in all, this work describes innovative methodological approaches for the identification of disease signatures through the combination of multi-omics analysis and provides novel knowledge on the pathological convergencies defining ALS.

Catanese Alberto, Rajkumar Sandeep, Sommer Daniel, Masrori Pegah, Hersmus Nicole, Van Damme Philip, Witzel Simon, Ludolph Albert, Ho Ritchie, Boeckers Tobias M, Mulaw Medhanie

2023-Mar-08

ALS, deep learning, motor neurons, omics

Dermatology Dermatology

A neutral comparison of statistical methods for analyzing longitudinally measured ordinal outcomes in rare diseases.

In Biometrical journal. Biometrische Zeitschrift

Ordinal data in a repeated measures design of a crossover study for rare diseases usually do not allow for the use of standard parametric methods, and hence, nonparametric methods should be considered instead. However, only limited simulation studies in settings with small sample sizes exist. Therefore, starting from an Epidermolysis Bullosa simplex trial with the above-mentioned design, a rank-based approach using the R package nparLD and different generalized pairwise comparisons (GPC) methods were compared impartially in a simulation study. The results revealed that there was not one single best method for this particular design, because a trade-off exists between achieving high power, accounting for period effects, and for missing data. Specifically, nparLD as well as the unmatched GPC approaches do not address crossover aspects, and the univariate GPC variants partly ignore the longitudinal information. The matched GPC approaches, on the other hand, take the crossover effect into account in the sense of incorporating the within-subject association. Overall, the prioritized unmatched GPC method achieved the highest power in the simulation scenarios, although this may be due to the specified prioritization. The rank-based approach yielded good power even at a sample size of N = 6 $N=6$ , whereas the matched GPC method could not control the type I error.

Geroldinger Martin, Verbeeck Johan, Thiel Konstantin E, Molenberghs Geert, Bathke Arne C, Laimer Martin, Zimmermann Georg

2023-Mar-08

Epidermolysis Bullosa simplex, generalized pairwise comparison (GPC), neutral comparison, nonparametric marginal model (nparLD), repeated measures

Radiology Radiology

Prediction of the sarcopenia in peritoneal dialysis using simple clinical information: A machine learning-based model.

In Seminars in dialysis

INTRODUCTION : Sarcopenia is associated with significant cardiovascular risk, and death in patients undergoing peritoneal dialysis (PD). Three tools are used for diagnosing sarcopenia. The evaluation of muscle mass requires dual energy X-ray absorptiometry (DXA) or computed tomography (CT), which is labor-intensive and relatively expensive. This study aimed to use simple clinical information to develop a machine learning (ML)-based prediction model of PD sarcopenia.

METHODS : According to the newly revised Asian Working Group for Sarcopenia (AWGS2019), patients were subjected to complete sarcopenia screening, including appendicular skeletal muscle mass, grip strength, and five-time chair stand time test. Simple clinical information such as general information, dialysis-related indices, irisin and other laboratory indices, and bioelectrical impedance analysis (BIA) data were collected. All data were randomly split into training (70%) and testing (30%) sets. Difference, correlation, univariate, and multivariate analyses were used to identify core features significantly associated with PD sarcopenia.

RESULT : 12 core features (C), namely, grip strength, body mass index (BMI), total body water value, irisin, extracellular water/total body water, fat-free mass index, phase angle, albumin/globulin, blood phosphorus, total cholesterol, triglyceride, and prealbumin were excavated for model construction. Two ML models, the neural network (NN), and support vector machine (SVM) were selected with tenfold cross-validation to determine the optimal parameter. The C-SVM model showed a higher area under the curve (AUC) of 0.82 (95% confidence interval [CI]: 0.67-1.00), with a highest specificity of 0.96, sensitivity of 0.91, positive predictive value (PPV) of 0.96, and negative predictive value (NPV) of 0.91.

CONCLUSION : The ML model effectively predicted PD sarcopenia and has clinical potential to be used as a convenient sarcopenia screening tool.

Wu Jiaying, Lin Shuangxiang, Guan Jichao, Wu Xiujuan, Ding Miaojia, Shen Shuijuan

2023-Mar-08

General General

ECG signal feature extraction trends in methods and applications.

In Biomedical engineering online

Signal analysis is a domain which is an amalgamation of different processes coming together to form robust pipelines for the automation of data analysis. When applied to the medical world, physiological signals are used. It is becoming increasingly common in today's day and age to be working with very large datasets, on the scale of having thousands of features. This is largely due to the fact that the acquisition of biomedical signals can be taken over multi-hour timeframes, which is another challenge to solve in and of itself. This paper will focus on the electrocardiogram (ECG) signal specifically, and common feature extraction techniques used for digital health and artificial intelligence (AI) applications. Feature extraction is a vital step of biomedical signal analysis. The basic goal of feature extraction is for signal dimensionality reduction and data compaction. In simple terms, this would allow one to represent data with a smaller subset of features; these features could then later be leveraged to be used more efficiently for machine learning and deep learning models for applications, such as classification, detection, and automated applications. In addition, the redundant data in the overall dataset is filtered out as the data is reduced during feature extraction. In this review, we cover ECG signal processing and feature extraction in the time domain, frequency domain, time-frequency domain, decomposition, and sparse domain. We also provide pseudocode for the methods discussed so that they can be replicated by practitioners and researchers in their specific areas of biomedical work. Furthermore, we discuss deep features, and machine learning integration, to complete the overall pipeline design for signal analysis. Finally, we discuss future work that can be innovated upon in the feature extraction domain for ECG signal analysis.

Singh Anupreet Kaur, Krishnan Sridhar

2023-Mar-08

Artificial intelligence, Digital health, ECG, Feature extraction, Signal analysis, Telehealth

General General

Application of machine learning to identify risk factors of birth asphyxia.

In BMC pregnancy and childbirth ; h5-index 58.0

BACKGROUND : Developing a prediction model that incorporates several risk factors and accurately calculates the overall risk of birth asphyxia is necessary. The present study used a machine learning model to predict birth asphyxia.

METHODS : Women who gave birth at a tertiary Hospital in Bandar Abbas, Iran, were retrospectively evaluated from January 2020 to January 2022. Data were extracted from the Iranian Maternal and Neonatal Network, a valid national system, by trained recorders using electronic medical records. Demographic factors, obstetric factors, and prenatal factors were obtained from patient records. Machine learning was used to identify the risk factors of birth asphyxia. Eight machine learning models were used in the study. To evaluate the diagnostic performance of each model, six metrics, including area under the receiver operating characteristic curve, accuracy, precision, sensitivity, specificity, and F1 score were measured in the test set.

RESULTS : Of 8888 deliveries, we identified 380 women with a recorded birth asphyxia, giving a frequency of 4.3%. Random Forest Classification was found to be the best model to predict birth asphyxia with an accuracy of 0.99. The analysis of the importance of the variables showed that maternal chronic hypertension, maternal anemia, diabetes, drug addiction, gestational age, newborn weight, newborn sex, preeclampsia, placenta abruption, parity, intrauterine growth retardation, meconium amniotic fluid, mal-presentation, and delivery method were considered to be the weighted factors.

CONCLUSION : Birth asphyxia can be predicted using a machine learning model. Random Forest Classification was found to be an accurate algorithm to predict birth asphyxia. More research should be done to analyze appropriate variables and prepare big data to determine the best model.

Darsareh Fatemeh, Ranjbar Amene, Farashah Mohammadsadegh Vahidi, Mehrnoush Vahid, Shekari Mitra, Jahromi Malihe Shirzadfard

2023-Mar-08

Birth asphyxia, Machine learning, Risk factors