Balanced Accuracy Values Research Articles

Radiomics-based machine learning role in differential diagnosis between small renal oncocytoma and clear cells carcinoma on contrast-enhanced CT: A pilot study

PurposeTo investigate the potential role of radiomics-based machine learning in differentiating small renal oncocytoma (RO) from clear cells carcinoma (ccRCC) on contrast-enhanced CT (CECT). Material and methodsFifty-two patients with small renal masses who underwent CECT before surgery between January 2016 and December 2020 were retrospectively included in the study. At pathology examination 39 ccRCC and 13 RO were identified. All lesions were manually delineated unenhanced (B), arterial (A) and venous (V) phases. Radiomics features were extracted using three different fixed bin widths (bw) of 25 HU, 10 HU, and 5 HU from each phase (B, A, V), and with different combinations (B+A, B+V, B+A+V, A+V), leading to 21 different datasets. Montecarlo Cross Validation technique was used to quantify the estimator performance. The final model built using the hyperparameter selected with Optuna was trained again on the training set and the final performance evaluation was made on the test set. ResultsThe A+V bw 10 achieved the greater median (IQR) balanced accuracy considering all the models of 0.70 (0.64–0.75), while A bw 10 considering only the monophasic ones. The A bw 10 model achieved a median (IQR) sensitivity of 0.60 (0.40–0.60), specificity of 0.80 (0.73–0.87), AUC-ROC of 0.77 (0.66–0.84), accuracy of 0.75 (0.70–0.80), and a Phi Coefficient of 0.38 (0.20–0.47). None of the nine models with the lowest mean balanced accuracy values implemented features from A. ConclusionThe A bw 10 model was identified as the most efficient mono-phasic model in differentiating small RO from ccRCC.

Improved classification of soil As contamination at continental scale: Resolving class imbalances using machine learning approach

The identification of arsenic (As)-contaminated areas is an important prerequisite for soil management and reclamation. Although previous studies have attempted to identify soil As contamination via machine learning (ML) methods combined with soil spectroscopy, they have ignored the rarity of As-contaminated soil samples, leading to an imbalanced learning problem. A novel ML framework was thus designed herein to solve the imbalance issue in identifying soil As contamination from soil visible and near-infrared spectra. Spectral preprocessing, imbalanced dataset resampling, and model comparisons were combined in the ML framework, and the optimal combination was selected based on the recall. In addition, Bayesian optimization was used to tune the model hyperparameters. The optimized model achieved recall, area under the curve, and balanced accuracy values of 0.83, 0.88, and 0.79, respectively, on the testing set. The recall was further improved to 0.87 with the threshold adjustment, indicating the model's excellent performance and generalization capability in classifying As-contaminated soil samples. The optimal model was applied to a global soil spectral dataset to predict areas at a high risk of soil As contamination on a global scale. The ML framework established in this study represents a milestone in the classification of soil As contamination and can serve as a valuable reference for contamination management in soil science.

Improving Stroke Detection with Hybrid Sampling and Cascade Generalization

The prevalence of stroke in Indonesia has increased. One survey in Indonesia that contains information about the health conditions of the Indonesian people is the Indonesian Family Life Survey (IFLS). The proportion of respondents who had a stroke and non-stroke in IFLS5 showed an imbalance with an extreme level of imbalance; hence, this research aims to overcome this problem with SMOTE, SMOTE-Tomek Link, and SMOTE-ENN; then, the balanced dataset is classified using the ensemble and cascade approaches to improve the detection of stroke risk and to identify the important variables. However, the stroke respondents were still challenging to classify after imbalance class handling, presumably because of the large amount of data before and after balancing. The solution is to balance the training data with various percentages. The results showed the best percentage is applied to 5% of the training data, balanced by the SMOTE-ENN, and the ensemble method with the cascade approach increases the sensitivity and balanced accuracy values. Random forest and logistic regression combine models that produce the best performance, with a classification tree as the final model. The important variables obtained from this combination are the addition of probability from random forest, logistic regression, history of hypertension, age, and physical activity.

Spanish vs USA cohort comparison of prehospital trauma scores to predict short-term mortality

BackgroundThis study aimed to evaluate three prehospital early warning scores (EWSs): RTS, MGAP and MREMS, to predict short-term mortality in acute life-threatening trauma and injury/illness by comparing United States (US) and Spanish cohorts. MethodsA total of 8,854 patients, 8,598/256 survivors/nonsurvivors, comprised the unified cohort. Datasets were randomly divided into training and test sets. Training sets were used to analyse the discriminative power of the scores in terms of the area under the curve (AUC), and the score performance was assessed in the test set in terms of sensitivity (SE), specificity (SP), accuracy (ACC) and balanced accuracy (BAC). ResultsThe three scores showed great discriminative power with AUCs>0.90, and no significant differences between cohorts were found. In the test set, RTS/MREMS/MGAP showed SE/SP/ACC/BAC values of 86.0/89.9/89.6/87.1%, 91.0/86.9/87.5/88.5%, and 87.7/82.9/83.4/85.2%, respectively. ConclusionsAll EWSs showed excellent ability to predict the risk of short-term mortality, independent of the country.

Knowledge Discovery from Confusion Matrix of Pruned CART in Imbalanced Microarray Data Ovarian Cancer Classification

Purpose: The results of microarray data analysis is important in cancer diagnosis, especially in early stages asymptomatic cancers like ovarian cancer. One of the challenges in analyzing microarray data is the problem of imbalanced data. Unfortunately, research that carries out cancer classification from microarray data often ignores this challenge, so that it doesn’t use appropriate evaluation metrics. It makes the results biased towards the majority class. This study uses a popular evaluation metric “accuracy” and an evaluation metric that is suitable for imbalanced data “balanced accuracy (BA)” to gain information from the confusion matrix regarding accuracy and BA values in case of ovarian cancer classification.Methods: This study use Classification and Regression Tree (CART) as the classifier. CART optimized by pruning. CART optimal is determined from the results of CART complexity analysis and confusion matrix.Results: The confusion matrix and CART interpretations in this research show that CART with low complexity is still able to predict majority class respondents well. However, when none of the data in the minority class was classified correctly, the accuracy value was still quite high, namely 86.97% and 88.03% respectively at the training and testing stages, while the BA value at both stages was only 50%.Novelty: It is very important to ensure that the evaluation metrics used match the characteristics of the data being processed. This research illustrate the difference between accuracy and BA. It concluded that that classification of an imbalanced dataset without doing resampling can use BA as evaluation metric, because based on the results, BA is more fairly to both classes.

Predicting S. aureus antimicrobial resistance with interpretable genomic space maps.

Increasing antimicrobial resistance (AMR) represents a global healthcare threat. To decrease the spread of AMR and associated mortality, methods for rapid selection of optimal antibiotic treatment are urgently needed. Machine learning (ML) models based on genomic data to predict resistant phenotypes can serve as a fast screening tool prior to phenotypic testing. Nonetheless, many existing ML methods lack interpretability. Therefore, we present a methodology for visualization of sequence space and AMR prediction based on the non-linear dimensionality reduction method - generative topographic mapping (GTM). This approach, applied to AMR data of >5000 S. aureus isolates retrieved from the PATRIC database, yielded GTM models with reasonable accuracy for all drugs (balanced accuracy values ≥0.75). The Generative Topographic Maps (GTMs) represent data in the form of illustrative maps of the genomic space and allow for antibiotic-wise comparison of resistant phenotypes. The maps were also found to be useful for the analysis of genetic determinants responsible for drug resistance. Overall, the GTM-based methodology is a useful tool for both the illustrative exploration of the genomic sequence space and AMR prediction.

Evaluating causative factors for landslide susceptibility along the Imphal-Jiribam railway corridor in the North-Eastern part of India using a GIS-based statistical approach.

The Northeast part of India is experiencing an increase in infrastructure projects as well as landslides. This study aims to prepare the landslide susceptibility map of Tamenglong and Senapati districts, Manipur, India, and evaluates the state of landslide susceptibility along the Imphal-Jiribam railway corridor. Efficient statistical methods such as frequency ratio (FR), information value (IoV), weight of evidence (WoE), and weighted linear combination (WLC) were used in model preparation. A total of 322 landslide points were randomly divided into training (70%) and testing (30%) datasets. Nine causative factors were utilized for landslide susceptibility mapping (LSM). The importance of which was obtained using the information gain (IG) method. FR, IoV, WoE, and WLC were used to prepare the LSM using the training datasets and nine causative factors. Moreover, the accuracy and consistency were evaluated using AUC-ROC, precision, recall, overall accuracy (OA), balanced accuracy (BA), and F-score. The validation results showed that all methods performed well with the highest AUC and precision values of 0.913 and 0.95, respectively, for the IoV method, while the WLC method had the highest OA, BA, and F-score values of 0.808, 0.81, and 0.812, respectively. Finally, the results from LSM were used to evaluate the state of landslide susceptibility along the Imphal-Jiribam railway corridor. The results showed that 34% of the areas had high and very high susceptibility, while 40% were under less and significantly less susceptibility. The Tupul landslide area lay in medium susceptibility where the disastrous landslide occurred on 30 June 2022. Susceptibility values around the Noney and Khongsag railway station ranged from high to very high susceptibility. Thus, the study manifests the need for LSM preparation in rapidly constructing areas, which in turn will help the policymakers and planners for adopting strategies to minimize losses caused due to landslides.

Modeling and insights into the structural characteristics of endocrine-disrupting chemicals

Endocrine-disrupting chemicals (EDCs) can cause serious harm to human health and the environment; therefore, it is important to rapidly and correctly identify EDCs. Different computational models have been proposed for the prediction of EDCs over the past few decades, but the reported models are not always easily available, and few studies have investigated the structural characteristics of EDCs. In the present study, we have developed a series of artificial intelligence models targeting EDC receptors: the androgen receptor (AR); estrogen receptor (ER); and pregnane X receptor (PXR). The consensus models achieved good predictive results for validation sets with balanced accuracy values of 87.37%, 90.13%, and 79.21% for AR, ER, and PXR binding assays, respectively. Analysis of the physical-chemical properties suggested that several chemical properties were significantly (p < 0.05) different between EDCs and non-EDCs. We also identified structural alerts that can indicate an EDC, which were integrated into the web server SApredictor. These models and structural characteristics can provide useful tools and information in the discrimination and mechanistic understanding of EDCs in drug discovery and environmental risk assessment.

Manifestation of depression in speech overlaps with characteristics used to represent and recognize speaker identity

The sound of a person’s voice is commonly used to identify the speaker. The sound of speech is also starting to be used to detect medical conditions, such as depression. It is not known whether the manifestations of depression in speech overlap with those used to identify the speaker. In this paper, we test the hypothesis that the representations of personal identity in speech, known as speaker embeddings, improve the detection of depression and estimation of depressive symptoms severity. We further examine whether changes in depression severity interfere with the recognition of speaker’s identity. We extract speaker embeddings from models pre-trained on a large sample of speakers from the general population without information on depression diagnosis. We test these speaker embeddings for severity estimation in independent datasets consisting of clinical interviews (DAIC-WOZ), spontaneous speech (VocalMind), and longitudinal data (VocalMind). We also use the severity estimates to predict presence of depression. Speaker embeddings, combined with established acoustic features (OpenSMILE), predicted severity with root mean square error (RMSE) values of 6.01 and 6.28 in DAIC-WOZ and VocalMind datasets, respectively, lower than acoustic features alone or speaker embeddings alone. When used to detect depression, speaker embeddings showed higher balanced accuracy (BAc) and surpassed previous state-of-the-art performance in depression detection from speech, with BAc values of 66% and 64% in DAIC-WOZ and VocalMind datasets, respectively. Results from a subset of participants with repeated speech samples show that the speaker identification is affected by changes in depression severity. These results suggest that depression overlaps with personal identity in the acoustic space. While speaker embeddings improve depression detection and severity estimation, deterioration or improvement in mood may interfere with speaker verification.

Analyzing the Effectiveness of Imbalanced Data Handling Techniques in Predicting Driver Phone Use

Distracted driving leads to a significant number of road crashes worldwide. Smartphone use is one of the most common causes of cognitive distraction among drivers. Available data on drivers’ phone use presents an invaluable opportunity to identify the main factors behind this behavior. Machine learning (ML) techniques are among the most effective techniques for this purpose. However, the potential and usefulness of these techniques are limited, due to the imbalance of available data. The majority class of instances collected is for drivers who do not use their phones, while the minority class is for those who do use their phones. This paper evaluates two main approaches for handling imbalanced datasets on driver phone use. These methods include oversampling and undersampling. The effectiveness of each method was evaluated using six ML techniques: Multilayer Perceptron (MLP), Support Vector Machine (SVM), Naive Bayes (NB), Bayesian Network (BayesNet), J48, and ID3. The proposed methods were also evaluated on three Deep Learning (DL) models: Arch1 (5 hidden layers), Arch2 (10 hidden layers), and Arch3 (15 hidden layers). The data used in this document were collected through a direct observation study to explore a set of human, vehicle, and road surface characteristics. The results showed that all ML methods, as well as DL methods, achieved balanced accuracy values for both classes. ID3, J48, and MLP methods outperformed the rest of the ML methods in all scenarios, with ID3 achieving slightly better accuracy. The DL methods also provided good performances, especially for the undersampling data. The results also showed that the classification methods performed best on the undersampled data. It was concluded that road classification has the highest impact on cell phone use, followed by driver age group, driver gender, vehicle type, and, finally, driver seatbelt usage.

Prediction of plasma trough concentration of voriconazole in adult patients using machine learning

ObjectivePlasma trough concentration of voriconazole (VCZ) was associated with its toxicity and efficacy. However, the nonlinear pharmacokinetic characteristics of VCZ make it difficult to determine the relationship between clinical characteristics and its concentration. We intended to present a machine learning (ML)-based method to predict toxic plasma trough concentration of VCZ (>5 μg/mL). MethodsA single center retrospective study was conducted. Three ML algorithms were used to estimate the concentration in adult patients, including random forest (RF), gradient boosting (GB), and extreme gradient boosting (XGBoost). The importance of variables was recognized by the SHapley Additive exPlanations (SHAP) method. In addition, an external validation set was used to validate the robustness of models. ResultsA total of 1318 VCZ plasma concentration were included, with 33 variables enrolled in the model. Nine classification models were developed using the RF, GB, and XGBoost algorithms. Most models performed well for both the training set and test set, with an average balanced accuracy (BA) of 0.704 and an average accuracy (ACC) of 0.788. In addition, the average Matthews correlation coefficient value reached 0.484, which indicated the predicted values are meaningful. Based on the average BA and ACC values, the predictive ability of the models can be ranked from best to worst as follows: younger adult models > mixed models > elderly models, and XGBoost models > GBT models > RF models. The SHAP results showed that the top five influencing factors in younger adult patients (<60 years) were albumin, total bile acid (TBA), platelets count, age, and inflammation, while the top five influencing factors in elderly patients were albumin, TBA, aspartate aminotransferase, creatinine, and alanine aminotransferase. Furthermore, the prediction of external validation set for VCZ concentrations verified the high reliability of the models, for the ACC value of 0.822 by the best model. ConclusionsThe ML models can be reliable tools for predicting toxic concentration exposure of VCZ. The SHAP results may provide useful guidelines for dosage adjustment of VCZ.

Ensemble machine learning algorithm for predicting acute kidney injury in patients admitted to the neurointensive care unit following brain surgery

Acute kidney injury (AKI) is a common postoperative complication among patients in the neurological intensive care unit (NICU), often resulting in poor prognosis and high mortality. In this retrospective cohort study, we established a model for predicting AKI following brain surgery based on an ensemble machine learning algorithm using data from 582 postoperative patients admitted to the NICU at the Dongyang People's Hospital from March 1, 2017, to January 31, 2020. Demographic, clinical, and intraoperative data were collected. Four machine learning algorithms (C5.0, support vector machine, Bayes, and XGBoost) were used to develop the ensemble algorithm. The AKI incidence in critically ill patients after brain surgery was 20.8%. Intraoperative blood pressure; postoperative oxygenation index; oxygen saturation; and creatinine, albumin, urea, and calcium levels were associated with the postoperative AKI occurrence. The area under the curve value for the ensembled model was 0.85. The accuracy, precision, specificity, recall, and balanced accuracy values were 0.81, 0.86, 0.44, 0.91, and 0.68, respectively, indicating good predictive ability. Ultimately, the models using perioperative variables exhibited good discriminatory ability for early prediction of postoperative AKI risk in patients admitted to the NICU. Thus, the ensemble machine learning algorithm may be a valuable tool for forecasting AKI.

Synthetic Minority Oversampling Technique Pada Model Logit dan Probit Status Pengangguran Terdidik

Educated unemployment is caused by a misalignment of educational development planning and employment development, resulting in underemployed graduates from various educational institutions. Unemployment data in DKI Jakarta shows an unequal class. Unbalanced data is a severe problem of modeling because it can cause prediction errors that affect the accuracy of the resulting model. Using SMOTE to handle unbalanced data will likely increase the model’s accuracy. This study aims to find the best model for identifying the factors influencing the status of educated unemployment using logit and probit models and handling unbalanced data using SMOTE. The results showed that the independent variables that affect the status of educated unemployment in the logit and probit models are the same: age group and participation in training. The independent variables that affect the status of educated unemployment in the logit and probit models with SMOTE are also the same: age group, marital status, and participation in training. Unbalanced data handling using SMOTE can increase the balanced accuracy value significantly. Balanced accuracy values for the logit and probit models with SMOTE are higher than the logit and probit models without SMOTE. The logit model with SMOTE is the best because it has the highest balanced accuracy value compared to other models. According to the logit model with SMOTE, the educated unemployed in DKI Jakarta are young and have never married. There is a need for the government to play a role in improving the quality of educational institutions in producing graduates who meet company qualifications and can be hired by employers. Unemployed people who have attended the training, despite having a higher education, may also become unemployed. The training provided has not been able to reduce the unemployment rate. As a result, the government should be able to provide training to improve entrepreneurship skills while also providing capital in the form of business loans to reduce educated unemployment.

Multi-class classification model for psychiatric disorder discrimination

BackgroundPhysicians follow-up a symptom-based approach in the diagnosis of psychiatric diseases. According to this approach, a process based on internationally valid diagnostic tools such as The Diagnostic and Statistical Manual of Mental Disorders (DSM) or International Classification of Diseases (ICD), patient reports and the observation and experience of the physician is monitored. As in other fields of medicine, the search for biomarkers that can be used in processes related to diseases continues in psychiatry and various researches are carried out in this field. ObjectivesWithin the scope of this study, a dataset containing electroencephalogram (EEG) measurements of individuals diagnosed with different psychiatric diseases were analyzed by machine learning methods and the diseases were differentiated/classified with the models obtained. Thus, it was investigated whether EEG data could be a biomarker for psychiatric diseases. Materials and methodsIn the dataset analyzed within the scope of the study, for 550 patients (81 bipolar disorder, 95 attention deficit and hyperactivity disorder - ADHD, 67 depression, 34 obsessive compulsive disorder - OCD, 75 opioid, 146 posttraumatic stress disorder - PTSD, 52 schizophrenia) and 84 healthy individuals, there are 634 samples (rows), 77 variables (columns) in total. 76 of the variables consist of absolute power values belonging to 4 frequency bands (alpha, beta, delta, theta) collected from 19 different electrodes. 80 % of the dataset was used for training the models and 20 % of the data was used for testing the performance of the models. The 5-fold cross validation (CV) method, which repeats 3 times in the training dataset, was used and with this method, the hyperparameters used in the models were also optimized. Different models have been established with the selected hyperparameters and the performance of these models has been tested with the test dataset. C5.0, random forest (RF), support vector machine (SVM) and artificial neural networks (ANN) were used to build the models. ResultsWithin the scope of the study, the absolute power values obtained from EEG measurements performed using 19 electrodes were analyzed by machine learning methods. It was concluded that classification between disease groups was feasible with a high accuracy (C5.0–0.841, SVM_radial − 0.841, RF − 0.762). It was observed that ADHD, depression and schizophrenia diseases can be differentiated better (F-score = 1, balanced accuracy = 1) once the results were evaluated on a class category basis according to the F- measure and balanced accuracy values. Discussion and conclusionThrough the medium of the analyzes made within the scope of this study, it was investigated whether EEG data could be used as a biomarker for the detection and diagnosis of psychiatric diseases. The findings obtained from this study revealed that by using EEG data as a biomarker, it can be highly predicted whether a person has a psychiatric disease or not. Once evaluated with broad strokes, it is feasible to assert that it is possible to analyze whether the person who consults a physician with a complaint is ranked among the psychiatric disease class with EEG measurement. When trying to differentiate between numerous and diverse disease categories, it may be claimed that some diseases (ADHD, depression, schizophrenia) can be distinguished better by coming to the fore on a model basis. Considering the findings, it is anticipated that the analyzes obtained as a result of this study will contribute to the studies to be conducted using machine learning in the field of psychiatry.

Predicting histopathology markers of endometrial carcinoma with a quantitative image analysis approach based on spherical harmonics in multiparametric MRI

PurposeIdentifying optimal machine learning pipelines for computer-aided diagnosis is key for the development of robust, reproducible, and clinically relevant imaging biomarkers for endometrial carcinoma. The purpose of this study was to introduce the mathematical development of image descriptors computed from spherical harmonics (SPHARM) decompositions as well as the associated machine learning pipeline, and to evaluate their performance in predicting deep myometrial invasion (MI) and histopathological high-grade in preoperative multiparametric magnetic resonance imaging (MRI). Patients and methodsThis retrospective study included 128 women with histopathology-confirmed endometrial carcinomas who underwent 1.5-T MRI before hysterectomy between January 2011 and July 2015. SPHARM descriptors of each tumor were computed on multiparametric MRI images (T2-weighted, diffusion-weighted, dynamic contrast-enhanced-MRI and apparent diffusion coefficient maps). Tensor-based logistic regression was used to classify two-dimensional SPHARM rotationally-invariant descriptors. Head-to-head comparisons with radiomics analyses were performed with DeLong tests with Bonferroni-Holm correction to compare diagnostic performances. ResultsWith all MRI contrasts, SPHARM analysis resulted in area under the curve, sensitivity, specificity, and balanced accuracy values of 0.94 (95% confidence interval [CI]: 0.85, 1.00), 100% (95% CI: 100, 100), 74% (95% CI: 51, 92), 87% (95% CI: 78, 98), respectively, for predicting deep MI. For predicting high-grade tumor histology, the corresponding values for the same diagnostic metrics were 0.81 (95% CI: 0.64, 0.90), 93% (95% CI: 67, 100), 63% (95% CI: 45, 79) and 78% (95% CI: 64, 86). The corresponding values achieved via radiomics were 0.92 (95% CI: 0.82, 0.95), 82% (95% CI: 65, 93), 80% (95% CI: 51, 94), 81% (95% CI: 70, 91) for deep MI and 0.72 (95% CI: 0.58, 0.83), 93% (95% CI: 65, 100), 55% (95% CI: 41, 69), 74% (95% CI: 52, 88) for high-grade histology. The diagnostic performance of the SPHARM analysis was not significantly different (P = 0.62) from that of radiomics for predicting deep MI but was significantly higher (P = 0.044) for predicting high-grade histology. ConclusionThe proposed SPHARM analysis yields similar or higher diagnostic performance than radiomics in identifying deep MI and high-grade status in histology-proven endometrial carcinoma.

A lightweight learning-based decoding algorithm for intraneural vagus nerve activity classification in pigs

Objective. Bioelectronic medicine is an emerging field that aims at developing closed-loop neuromodulation protocols for the autonomic nervous system (ANS) to treat a wide range of disorders. When designing a closed-loop protocol for real time modulation of the ANS, the computational execution time and the memory and power demands of the decoding step are important factors to consider. In the context of cardiovascular and respiratory diseases, these requirements may partially explain why closed-loop clinical neuromodulation protocols that adapt stimulation parameters on patient’s clinical characteristics are currently missing. Approach. Here, we developed a lightweight learning-based decoder for the classification of cardiovascular and respiratory functional challenges from neural signals acquired through intraneural electrodes implanted in the cervical vagus nerve (VN) of five anaesthetized pigs. Our algorithm is based on signal temporal windowing, nine handcrafted features, and random forest (RF) model for classification. Temporal windowing ranging from 50 ms to 1 s, compatible in duration with cardio-respiratory dynamics, was applied to the data in order to mimic a pseudo real-time scenario. Main results. We were able to achieve high balanced accuracy (BA) values over the whole range of temporal windowing duration. We identified 500 ms as the optimal temporal windowing duration for both BA values and computational execution time processing, achieving more than 86% for BA and a computational execution time of only ∼6.8 ms. Our algorithm outperformed in terms of BA and computational execution time a state of the art decoding algorithm tested on the same dataset (Vallone et al 2021 J. Neural Eng. 18 0460a2). We found that RF outperformed other machine learning models such as support vector machines, K-nearest neighbors, and multi-layer perceptrons. Significance. Our approach could represent an important step towards the implementation of a closed-loop neuromodulation protocol relying on a single intraneural interface able to perform real-time decoding tasks and selective modulation of the VN.

Prenatal prediction of neonatal respiratory morbidity: a radiomics method based on imbalanced few-shot fetal lung ultrasound images

BackgroundTo develop a non-invasive method for the prenatal prediction of neonatal respiratory morbidity (NRM) by a novel radiomics method based on imbalanced few-shot fetal lung ultrasound images.MethodsA total of 210 fetal lung ultrasound images were enrolled in this study, including 159 normal newborns and 51 NRM newborns. Fetal lungs were delineated as the region of interest (ROI), where radiomics features were designed and extracted. Integrating radiomics features selected and two clinical features, including gestational age and gestational diabetes mellitus, the prediction model was developed and evaluated. The modelling methods used were data augmentation, cost-sensitive learning, and ensemble learning. Furthermore, two methods, which embed data balancing into ensemble learning, were employed to address the problems of imbalance and few-shot simultaneously.ResultsOur model achieved sensitivity values of 0.82, specificity values of 0.84, balanced accuracy values of 0.83 and area under the curve values of 0.87 in the test set. The radiomics features extracted from the ROIs at different locations within the lung region achieved similar classification performance outcomes.ConclusionThe feature set we designed can efficiently and robustly describe fetal lungs for NRM prediction. RUSBoost shows excellent performance compared to state-of-the-art classifiers on the imbalanced few-shot dataset. The diagnostic efficacy of the model we developed is similar to that of several previous reports of amniocentesis and can serve as a non-invasive, precise evaluation tool for NRM prediction.

Hyperspectral Imaging Combined with Machine Learning for the Detection of Fusiform Rust Disease Incidence in Loblolly Pine Seedlings

Loblolly pine is an economically important timber species in the United States, with almost 1 billion seedlings produced annually. The most significant disease affecting this species is fusiform rust, caused by Cronartium quercuum f. sp. fusiforme. Testing for disease resistance in the greenhouse involves artificial inoculation of seedlings followed by visual inspection for disease incidence. An automated, high-throughput phenotyping method could improve both the efficiency and accuracy of the disease screening process. This study investigates the use of hyperspectral imaging for the detection of diseased seedlings. A nursery trial comprising families with known in-field rust resistance data was conducted, and the seedlings were artificially inoculated with fungal spores. Hyperspectral images in the visible and near-infrared region (400–1000 nm) were collected six months after inoculation. The disease incidence was scored with traditional methods based on the presence or absence of visible stem galls. The seedlings were segmented from the background by thresholding normalized difference vegetation index (NDVI) images, and the delineation of individual seedlings was achieved through object detection using the Faster RCNN model. Plant parts were subsequently segmented using the DeepLabv3+ model. The trained DeepLabv3+ model for semantic segmentation achieved a pixel accuracy of 0.76 and a mean Intersection over Union (mIoU) of 0.62. Crown pixels were segmented using geometric features. Support vector machine discrimination models were built for classifying the plants into diseased and non-diseased classes based on spectral data, and balanced accuracy values were calculated for the comparison of model performance. Averaged spectra from the whole plant (balanced accuracy = 61%), the crown (61%), the top half of the stem (77%), and the bottom half of the stem (62%) were used. A classification model built using the spectral data from the top half of the stem was found to be the most accurate, and resulted in an area under the receiver operating characteristic curve (AUC) of 0.83.

Comparing the Predictivity of Human Placental Gene, microRNA, and CpG Methylation Signatures in Relation to Perinatal Outcomes.

Molecular signatures are being increasingly integrated into predictive biology applications. However, there are limited studies comparing the overall predictivity of transcriptomic versus epigenomic signatures in relation to perinatal outcomes. This study set out to evaluate mRNA and microRNA (miRNA) expression and cytosine-guanine dinucleotide (CpG) methylation signatures in human placental tissues and relate these to perinatal outcomes known to influence maternal/fetal health; namely, birth weight, placenta weight, placental damage, and placental inflammation. The following hypotheses were tested: (1) different molecular signatures will demonstrate varying levels of predictivity towards perinatal outcomes, and (2) these signatures will show disruptions from an example exposure (ie, cadmium) known to elicit perinatal toxicity. Multi-omic placental profiles from 390 infants in the Extremely Low Gestational Age Newborns cohort were used to develop molecular signatures that predict each perinatal outcome. Epigenomic signatures (ie, miRNA and CpG methylation) consistently demonstrated the highest levels of predictivity, with model performance metrics including R2 (predicted vs observed) values of 0.36-0.57 for continuous outcomes and balanced accuracy values of 0.49-0.77 for categorical outcomes. Top-ranking predictors included miRNAs involved in injury and inflammation. To demonstrate the utility of these predictive signatures in screening of potentially harmful exogenous insults, top-ranking miRNA predictors were analyzed in a separate pregnancy cohort and related to cadmium. Key predictive miRNAs demonstrated altered expression in association with cadmium exposure, including miR-210, known to impact placental cell growth, blood vessel development, and fetal weight. These findings inform future predictive biology applications, where additional benefit will be gained by including epigenetic markers.

Bayesian network models with decision tree analysis for management of childhood malaria in Malawi

BackgroundMalaria is a major cause of death in children under five years old in low- and middle-income countries such as Malawi. Accurate diagnosis and management of malaria can help reduce the global burden of childhood morbidity and mortality. Trained healthcare workers in rural health centers manage malaria with limited supplies of malarial diagnostic tests and drugs for treatment. A clinical decision support system that integrates predictive models to provide an accurate prediction of malaria based on clinical features could aid healthcare workers in the judicious use of testing and treatment. We developed Bayesian network (BN) models to predict the probability of malaria from clinical features and an illustrative decision tree to model the decision to use or not use a malaria rapid diagnostic test (mRDT).MethodsWe developed two BN models to predict malaria from a dataset of outpatient encounters of children in Malawi. The first BN model was created manually with expert knowledge, and the second model was derived using an automated method. The performance of the BN models was compared to other statistical models on a range of performance metrics at multiple thresholds. We developed a decision tree that integrates predictions with the costs of mRDT and a course of recommended treatment.ResultsThe manually created BN model achieved an area under the ROC curve (AUC) equal to 0.60 which was statistically significantly higher than the other models. At the optimal threshold for classification, the manual BN model had sensitivity and specificity of 0.74 and 0.42 respectively, and the automated BN model had sensitivity and specificity of 0.45 and 0.68 respectively. The balanced accuracy values were similar across all the models. Sensitivity analysis of the decision tree showed that for values of probability of malaria below 0.04 and above 0.40, the preferred decision that minimizes expected costs is not to perform mRDT.ConclusionIn resource-constrained settings, judicious use of mRDT is important. Predictive models in combination with decision analysis can provide personalized guidance on when to use mRDT in the management of childhood malaria. BN models can be efficiently derived from data to support clinical decision making.