Supervised Machine Learning Methods Research Articles

Risk Characterization of Firms with ESG Attributes Using a Supervised Machine Learning Method

We examine the risk–return tradeoff of a portfolio of firms that have tangible environmental, social, and governance (ESG) attributes. We introduce a new type of penalized regression using the Mahalanobis distance-based method and show its usefulness using our sample of ESG firms. Our results show that ESG companies are exposed to financial state variables that capture the changes in investment opportunities. However, we find that there is no economically significant difference between the risk-adjusted returns of various ESG-rating-based portfolios and that the risk associated with a poor ESG rating portfolio is not significantly different than that of a good ESG rating portfolio. Although investors require return compensation for holding ESG stocks, the fact that the risk of a poor ESG rating portfolio is comparable to that of a good ESG rating portfolio suggests risk dimensions that go beyond ESG attributes. We further show that the new covariance-adjusted penalized regression improves the out-of-sample cross-sectional predictions of the ESG portfolio's expected returns. Overall, our approach is pragmatic and based on the ease of an empirical appeal.

DETECTION OF WIRELESS NETWORK ATTACKS USING SUPERVISED MACHINE LEARNING TECHNIQUE

The security and integrity of computer networks are seriously threatened by network assaults. Keeping a safe network environment requires the capacity to anticipate and stop these threats. Supervised machine learning methods have become powerful instruments for attacking network traffic and spotting patterns that point to malicious behaviour. We provide an in-depth examination of supervised machine learning methods for network attack prediction. We gather the data, preprocess it, extract pertinent features, and structure it so that machine learning algorithms may use it.We assess these algorithms' performance. To understand the fundamental patterns and traits of network assaults, we look at how interpretability the trained models are. This enables network managers to comprehend the types of threats and create suitable. Keywords - Network Attacks, Supervised Machine Learning, Network security,Network traffic analysis, Malicious behavior detection, Feature extraction,Machine learning algorithm,Model interpretability,Network attack patterns,Network defense strategies,Naive Bayes Algorithm, Ridge Classfier ,Random Forest Classfier

Application of Machine Learning Algorithms for Prediction of Tumor T-Cell Immunogens

The identification and characterization of immunogenic tumor antigens are essential for cancer vaccine development. In light of the impracticality of isolating and evaluating each putative antigen individually, in silico prediction algorithms, particularly those utilizing machine learning (ML) approaches, play a pivotal role. These algorithms significantly reduce the experimental workload necessary for discovering vaccine candidates. In this study, we employed six supervised ML methods on a dataset comprising 212 experimentally validated human tumor peptide antigens and an equal number of non-antigenic human peptides to develop models for immunogenicity prediction. These methods encompassed k-nearest neighbor (kNN), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost). The models underwent validation through internal cross-validation within 10 groups from the training set and were further assessed using an external test set. Remarkably, the kNN model demonstrated superior performance, recognizing 90% of the known immunogens in the test set. The RF model excelled in the identification of non-immunogens, accurately classifying 93% of them in the test set. The three top-performing ML models according to multiple evaluation metrics (SVM, RF, and XGBoost) are to be subsequently integrated into the new version of the VaxiJen server, facilitating tumor antigen prediction through a majority voting mechanism.

Information content of JWST spectra of WASP-39b

The era of James Webb Space Telescope (JWST) transmission spectroscopy of exoplanetary atmospheres commenced with the study of the Saturn-mass gas giant WASP-39b as part of the Early Release Science (ERS) program. WASP-39b was observed using several different JWST instrument modes (NIRCam, NIRISS, NIRSpec G395H and NIRSpec PRISM) and the spectra were published in a series of papers by the ERS team. The current study examines the information content of these spectra measured using the different instrument modes, focusing on the complexity of the temperature-pressure profiles and number of chemical species warranted by the data. We examine if the molecules H2O CO CO2 K H2S CH4 and SO2 are detected in each of the instrument modes. Two Bayesian inference methods are used to perform atmospheric retrievals: the standard nested sampling method, as well as the supervised machine learning method of the random forest (trained on a model grid). For nested sampling, Bayesian model comparison is used as a guide to identify the set of models with the required complexity to explain the data. Generally, non-isothermal transit chords are needed to fit the transmission spectra of WASP-39b, although the complexity of the temperature-pressure profile required is mode-dependent. The minimal set of chemical species needed to fit a spectrum is mode-dependent as well, and also depends on whether grey or non-grey clouds are assumed. When a non-grey cloud model is used to fit the NIRSpec G395H spectrum, it generates a spectral continuum that compensates for the water opacity. The same compensation is absent when fitting the non-grey cloud model to the NIRSpec PRISM spectrum (which has broader wavelength coverage), suggesting that it is spurious. The interplay between the cloud spectral continuum and the water opacity determines if sulphur dioxide is needed to fit either spectrum. The inferred elemental abundances of carbon and oxygen and the carbon-to-oxygen (C/O) ratios are all mode- and model-dependent, and should be interpreted with caution. Bayesian model comparison does not always offer a clear path forward for favouring specific retrieval models (e.g. grey versus non-grey clouds) and thus for enabling unambiguous interpretations of exoplanet spectra.

Machine Learning-assisted SERS Approach Enables the Biochemical Discrimination in Bcl-2 and Mcl-1 Expressing Yeast Cells Treated with Ketoconazole and Fluconazole Antifungals

Antifungal medications are important due to their potential application in cancer treatment either on their own or with traditional treatments. The mechanisms that prevent the effects of these medications and restrict their usage in cancer treatment are not completely understood. The evaluation and discrimination of the possible protective effects of the anti-apoptotic members of the Bcl-2 family of proteins, critical regulators of mitochondrial apoptosis, against antifungal drug-induced cell death has still scientific uncertainties that must be considered. Novel, simple, and reliable strategies are highly demanded to identify the biochemical signature of this phenomenon. However, the complex nature of cells poses challenges for the analysis of cellular biochemical changes or classification. In this study, for the first time, we investigated the probable protective activities of Bcl-2 and Mcl-1 proteins against cell damage induced by ketoconazole (KET) and fluconazole (FLU) antifungal drugs in a yeast model through surface-enhanced Raman spectroscopy (SERS) approach. The proposed SERS platform created robust Raman spectra with a high signal-to-noise ratio. The analysis of SERS spectral data via advanced unsupervised and supervised machine learning methods enabled unquestionable differentiation (100%) in samples and biomolecular identification. Various SERS bands related to lipids and proteins observed in the analyses suggest that the expression of these anti-apoptotic proteins reduces oxidative biomolecule damage induced by the antifungals. Also, cell viability assay, Annexin V-FITC/PI double staining, and total oxidant and antioxidant status analyses were performed to support Raman measurements. We strongly believe that the proposed approach paves the way for the evaluation of various biochemical structures/changes in various cells.

WBGT prediction with high spatial resolution using actual measurement data and data acquired using infrared sensors mounted on UAVs

Spatial understanding of heat stress indices is important to prevent the worsening of heat-related health problems. The objective of this study is to understand heat stress indices with high spatial and temporal resolution using actual measurement data and data acquired using infrared sensors mounted on unmanned aerial vehicles (UAVs). Two study sites were established to examine the relationship between green space and heat stress indices and the versatility of the developed method. The wet bulb globe temperature (WBGT), a heat stress index, was measured on extremely hot days. Thermal infrared, the normalized difference vegetation index, and the normalized difference water index were obtained from UAV images. Extreme gradient boosting, a supervised machine learning method for gradient boosting decision trees, was used to predict the WBGT for the study area using data from actual measurement points as training data. The maximum and average root mean square errors of the predicted WBGT were 2.8 and 1.4, respectively. The second measurement session, which was closest to noon, showed a greater WBGT reduction effect because of vegetation and moisture. This study successfully predicted the WBGT with high spatial and temporal resolution. This achievement is expected to contribute to the prevention of heat stress.

Narcissus reflected: Grey and white matter features joint contribution to the default mode network in predicting narcissistic personality traits.

Despite the clinical significance of narcissistic personality, its neural bases have not been clarified yet, primarily because of methodological limitations of the previous studies, such as the low sample size, the use of univariate techniques and the focus on only one brain modality. In this study, we employed for the first time a combination of unsupervised and supervised machine learning methods, to identify the joint contributions of grey matter (GM) and white matter (WM) to narcissistic personality traits (NPT). After preprocessing, the brain scans of 135 participants were decomposed into eight independent networks of covarying GM and WM via parallel ICA. Subsequently, stepwise regression and Random Forest were used to predict NPT. We hypothesizedthat a fronto-temporo parietal network, mainly related to the default mode network, may be involved in NPT and associated WM regions. Results demonstrated a distributed network that included GM alterations in fronto-temporal regions, the insula and the cingulate cortex, along with WM alterations in cerebellar and thalamic regions. To assess the specificity of our findings, we also examined whether the brain network predicting narcissism could also predict other personality traits (i.e., histrionic, paranoid and avoidant personalities). Notably, this network did not predict such personality traits. Additionally, a supervised machine learning model (Random Forest) was used to extract a predictive model for generalization to new cases. Results confirmed that the same network could predict new cases. These findings hold promise for advancing our understanding of personality traits and potentially uncovering brain biomarkers associated with narcissism.

Machine learning-based prediction of heat transfer performance in annular fins with functionally graded materials

This paper presents a study investigating the performance of functionally graded material (FGM) annular fins in heat transfer applications. An annular fin is a circular or annular structure used to improve heat transfer in various systems such as heat exchangers, electronic cooling systems, and power generation equipment. The main objective of this study is to analyze the efficiency of the ring fin in terms of heat transfer and temperature distribution. The fin surfaces are exposed to convection and radiation to dissipate heat. A supervised machine learning method was used to study the heat transfer characteristics and temperature distribution in the annular fin. In particular, a feedback architecture with the BFGS Quasi-Newton training algorithm (trainbfg) was used to analyze the solutions of the mathematical model governing the problem. This approach allows an in-depth study of the performance of fins, taking into account various physical parameters that affect its performance. To ensure the accuracy of the obtained solutions, a comparative analysis was performed using guided machine learning. The results were compared with those obtained by conventional methods such as the homotopy perturbation method, the finite difference method, and the Runge–Kutta method. In addition, a thorough statistical analysis was performed to confirm the reliability of the solutions. The results of this study provide valuable information on the behavior and performance of annular fins made from functionally graded materials. These findings contribute to the design and optimization of heat transfer systems, enabling better heat management and efficient use of available space.

Enhancing selection of alcohol consumption-associated genes by random forest.

Machine learning methods have been used in identifying omics markers for a variety of phenotypes. We aimed to examine whether a supervised machine learning algorithm can improve identification of alcohol-associated transcriptomic markers. In this study, we analysed array-based, whole-blood derived expression data for 17873 gene transcripts in 5508 Framingham Heart Study participants. By using the Boruta algorithm, a supervised random forest (RF)-based feature selection method, we selected twenty-five alcohol-associated transcripts. In a testing set (30 % of entire study participants), AUC (area under the receiver operating characteristics curve) of these twenty-five transcripts were 0·73, 0·69 and 0·66 for non-drinkers v. moderate drinkers, non-drinkers v. heavy drinkers and moderate drinkers v. heavy drinkers, respectively. The AUC of the selected transcripts by the Boruta method were comparable to those identified using conventional linear regression models, for example, AUC of 1958 transcripts identified by conventional linear regression models (false discovery rate < 0·2) were 0·74, 0·66 and 0·65, respectively. With Bonferroni correction for the twenty-five Boruta method-selected transcripts and three CVD risk factors (i.e. at P < 6·7e-4), we observed thirteen transcripts were associated with obesity, three transcripts with type 2 diabetes and one transcript with hypertension. For example, we observed that alcohol consumption was inversely associated with the expression of DOCK4, IL4R, and SORT1, and DOCK4 and SORT1 were positively associated with obesity, and IL4R was inversely associated with hypertension. In conclusion, using a supervised machine learning method, the RF-based Boruta algorithm, we identified novel alcohol-associated gene transcripts.

A mathematical model of competition between fiber and mucin degraders in the gut provides a possible explanation for mucus thinning

The human gut microbiota relies on complex carbohydrates (glycans) for energy and growth, primarily dietary fiber and host-derived mucins. We introduce a mathematical model of a glycan generalist and a mucin specialist in a two-compartment chemostat model of the human colon. Our objective is to characterize the influence of dietary fiber and mucin supply on the abundance of mucin-degrading species within the gut ecosystem. Current mathematical gut reactor models that include the enzymatic degradation of glycans do not differentiate between glycan types and their degraders. The model we present distinguishes between a generalist that can degrade both dietary fiber and mucin, and a specialist species that can only degrade mucin. The integrity of the colonic mucus barrier is essential for overall human health and well-being, with the mucin specialist Akkermanisa muciniphila being associated with a healthy mucus layer. Competition, particularly between the specialist and generalists like Bacteroides thetaiotaomicron, may lead to mucus layer erosion, especially during periods of dietary fiber deprivation. Our model treats the colon as a gut reactor system, dividing it into two compartments that represent the lumen and the mucus of the gut, resulting in a complex system of ordinary differential equations with a large and uncertain parameter space. To understand the influence of model parameters on long-term behavior, we employ a random forest classifier, a supervised machine learning method. Additionally, a variance-based sensitivity analysis is utilized to determine the sensitivity of steady-state values to changes in model parameter inputs. By constructing this model, we can investigate the underlying mechanisms that control gut microbiota composition and function, free from confounding factors.

A comprehensive framework for predicting electric vehicle's participation in ancillary service markets

AbstractElectric vehicles (EVs) have significant potential to offer unused capacity in ancillary service markets, providing unique opportunities for market operators to utilise these resources. EVs have a rapid response and high availability, making them a good fit for the frequency containment reserve (FCR) market. However, EV aggregators (EVAGs) must aggregate capacity blocks due to the limited capacity of individual EVs. An application of a supervised machine learning method named XGBoost is suggested to help EVAGs predict the amount of EV participation in the FCR market. The objective is to forecast yearly involvement using data from only a single week, using the game theory method SHapley Additive exPlanations (SHAP) to minimise extra data. The proposed strategy helps aggregators and uses feature engineering to select EVs with high potential to boost revenue. The proposed framework is effective in predicting EV performance in the DK‐2 market, as shown by multiple analyses.

Maximum Entropy Theory of Multiscale Coarse-Graining via Matching Thermodynamic Forces: Application to a Molecular Crystal (TATB).

The MSCG/FM (multiscale coarse-graining via force-matching) approach is an efficient supervised machine learning method to develop microscopically informed coarse-grained (CG) models. We present a theory based on the principle of maximum entropy (PME) enveloping the existing MSCG/FM approaches. This theory views the MSCG/FM method as a special case of matching the thermodynamic forces from the extended ensemble described by the set of thermodynamic (relevant) system coordinates. This set may include CG coordinates, the stress tensor, applied external fields, and so forth, and may be characterized by nonequilibrium conditions. Following the presentation of the theory, we discuss the consistent matching of both bonded and nonbonded interactions. The proposed PME formulation is used as a starting point to extend the MSCG/FM method to the constant strain ensemble, which together with the explicit matching of the bonded forces is better suited for coarse-graining anisotropic media at a submolecular resolution. The theory is demonstrated by performing the fine coarse-graining of crystalline 1,3,5-triamino-2,4,6-trinitrobenzene (TATB), a well-known insensitive molecular energetic material, which exhibits highly anisotropic mechanical properties.

Sentiment Classification of Post-Earthquake Consumer Brand Hate on Social Media Using Machine Learning Techniques

The widespread use of social media allows consumers to evaluate brands and to get into a direct interaction with brands and other followers of the same brands. After the devastating earthquake on February 6th, 2023, in ten provinces in Turkey a social media brand hatred was observed on two global brands Netflix and Starbucks. Brands were accused of not showing the necessary sensitivity and empathy towards the affected and the brand devotees. The objective of this study is to examine and classify brand hatred in online consumer-generated content using supervised machine learning methods. While the construct of brand hate has been extensively investigated in the discipline of marketing using different data collection methodologies, this is one of the first attempts to use machine learning methods for the analysis of the phenomenon. Unlike classic polarization, the labeling process was associated with the size of brand hatred; 0 denotes neutral reactions, -1 negative emotional reactions, and -2 negative relationship reactions. Support Vector Machines (SVM) was identified as the most successful algorithm for the explanation of the phenomenon.

Machine learning-based models to predict aquatic ecological risk for engineered nanoparticles: using hazard concentration for 5% of species as an endpoint.

Assessment and prediction for the ecotoxicity of engineered nanoparticles (ENPs) at the community or ecosystem levels represents a critical step toward a comprehensive understanding of the ecological risks of ENPs. Current studies on predicting the ecotoxicity of ENPs primarily focus on the cellular and individual levels, with limited exploration at the community or ecosystem levels. Herein, we present the first of the reports for the direct prediction of aquatic ecological risk for ENPs at the community level using machine learning (ML) approaches in the field of computational toxicology. Specifically, we extensively collected the threshold concentrations of twelve ENPs including metal- and carbon-based nanoparticles for aquatic species, i.e., hazardous concentrations at which 5% of species are harmed (HC5), established by a species sensitivity distribution. Afterwards, we used eight supervised ML methods including Adaboost, artificial neural network, C4.5 decision tree, K-nearest neighbor, logistic regression, Naive Bayes, random forest, and support vector machine to develop nine classification models and four regression models, respectively, for the qualitative and quantitative prediction of HC5. The evaluation of model performance yielded the internal validation accuracy of all classification models ranging from 71.4 to 100%, and the determination coefficient of regression models ranging from 0.702 to 0.999, indicating that the developed models showed good performance. By using a cross-validation method and an application domain characterization, the selected models were further validated to have powerful predictive ability. Furthermore, the incorporation of three nanostructural descriptors (metal oxide sublimation enthalpy, zeta potential, and specific surface area) linked to toxicity mechanisms (the release of metal ions, the stability of dispersions of particles in aqueous suspensions, and the surface properties of the material) effectively enhanced the prediction power and mechanistic interpretability of the selected models. These findings would not only be beneficial in the screening of ENPs with potential high ecological risks that need to be tested as a priority but also contribute to the development of environmental regulations and standards for ENPs.

Is the expression of different discrete emotions related to time? Evidence from online Chinese reviews using sentiment analysis and human behavior dynamics.

The online behavior of online users has taken on complex and diverse characteristics, and posting product reviews on e-commerce platforms is no exception. In fact, reviews contain rich and multi-dimensional discrete emotional information, and whether there is a relationship between the expression of these different discrete emotions and the time interval between product purchase and review posting as well as their related characteristics are the issues that this study needs to analyze and solve in depth. Based on the OCC model (named after three proposers) of psychological emotional cognitive evaluation theory as the basis for emotion classification, the study used the massive amounts of Chinese reviews of mobile phones on the Chinese e-commerce platform Jingdong Mall as the research object, applied supervised machine learning methods to classify discrete emotions, and constructed a large corpus containing satisfaction, disappointment, admiration, reproach, love, and hate; then the study delved into the distribution and behavioral dynamics characteristics of consumers' comments containing the different discrete emotions at different "purchase-comment" time intervals. The results showed that the first peak of the distribution curves of the six discrete emotions at different "purchase-comment" time intervals occurs on the first day after purchase and then decreases gradually but at different rates. The three curves for satisfaction, love, and hate emotions also show a second peak on the eleventh day, which is more similar to the bimodal distribution, implying that the corresponding product reviews are more objective. In addition, the distribution of reviews containing the six discrete emotions at different "purchase-comment" time intervals follows a power-law distribution and has the temporal characteristics of human behavioral dynamics, that is, "strong paroxysms and weak memory". However, the reviews containing the admiration and reproach emotions were most intensively written by consumers after the purchase, indicating that the service provided by the seller, logistics, and e-commerce platform stimulates more consumers to give quick responses and detailed reviews. This study is not only of great significance for exploring the internal mechanisms of consumer discrete emotional expression but also provides important decision-making references for potential consumer purchasing decisions, product updates for developers, marketing strategy formulation for marketing teams, and service improvement for sellers, logistics companies, and e-commerce platforms.

A NOVEL CORONARY HEART STROKE PREDICTION SYSTEM USING MACHINE LEARNING TECHNIQUES

Over the past few decades, cardiovascular diseases have surpassed all other causes of death as the main killers in industrialised, underdeveloped, and developing nations. Early detection of heart conditions and clinical care can lower the death rate. Based on the patient's various cardiac features, we proposed a model for forecasting heart disease and identifying impending heart disease using machine learning techniques In most cases,input is received through numerical data of various parameters, and output findings are generated in real-time, predicting whether or notthe patient has a disease. We'll use a variety of supervised machine learning methods before deciding which one is best for the model. Existing systems rely on classical deep learning models, which are inefficient and imprecise. They aren't as accurate as the proposed model and take a little longer to process.

Predicting the risk of becoming eligible for the disability pension: Machine learning methods applied to French health data

Benefiting from the disability pension implies morbid (physical and psychological) and social (fall in income) implications for the person. It also has economic consequences for society, with increasing expenses since 2011 (+4.9% on average per year). Investing in preventive actions against the loss of the ability to work should limit these consequences, but it requires targeting people at risk. The development of artificial intelligence opens up prospects in this regard. To target, using supervised machine learning methods, those people with a high probability of becoming eligible for the disability pension over the course of the year based on their socio-demographic and medical characteristics (pathologies, work stoppages, drugs taken, and medical procedures). Among the beneficiaries of the French public welfare system aged 20–64 in 2017, we compared the socio-demographic and medical characteristics between 2014 and 2016 of those who received a disability pension in 2017 and not before, and those who did not receive a disability pension from 2014 to 2017. The determination of the boundary between these two groups was tested using logistic regression, decision trees, random forests, naive Bayes classifiers, and support vector machines. The models’ performance was compared with respect to accuracy, precision, sensitivity, specificity, and AUC (area under the curve). Finally, the predictive power of each factor was measured by AUC too. The boosted logistic regression had the best performance for three of the five criteria, but low sensitivity. The best sensitivity was obtained with the support vector machines, with an accuracy close to that of the boosted logistic regression, but a lower precision and specificity. Random forests offered the best discriminatory ability. The naive Bayes classifier had the worst performance. The most predictive factors in becoming eligible for the disability pension were having 30 days or more off sick in 2014, 2015, and 2016 and being aged 55 to 64. Supervised learning methods have appeared relevant for identifying people with the highest probability of becoming eligible for the disability pension and, more broadly, for steering public and social policies.

Supervised Parametric Learning in the Identification of Composite Biomarker Signatures of Type 1 Diabetes in Integrated Parallel Multi-Omics Datasets.

Type 1 diabetes (T1D) is a devastating autoimmune disease, and its rising prevalence in the United States and around the world presents a critical problem in public health. While some treatment options exist for patients already diagnosed, individuals considered at risk for developing T1D and who are still in the early stages of their disease pathogenesis without symptoms have no options for any preventive intervention. This is because of the uncertainty in determining their risk level and in predicting with high confidence who will progress, or not, to clinical diagnosis. Biomarkers that assess one's risk with high certainty could address this problem and will inform decisions on early intervention, especially in children where the burden of justifying treatment is high. Single omics approaches (e.g., genomics, proteomics, metabolomics, etc.) have been applied to identify T1D biomarkers based on specific disturbances in association with the disease. However, reliable early biomarkers of T1D have remained elusive to date. To overcome this, we previously showed that parallel multi-omics provides a more comprehensive picture of the disease-associated disturbances and facilitates the identification of candidate T1D biomarkers. This paper evaluated the use of machine learning (ML) using data augmentation and supervised ML methods for the purpose of improving the identification of salient patterns in the data and the ultimate extraction of novel biomarker candidates in integrated parallel multi-omics datasets from a limited number of samples. We also examined different stages of data integration (early, intermediate, and late) to assess at which stage supervised parametric models can learn under conditions of high dimensionality and variation in feature counts across different omics. In the late integration scheme, we employed a multi-view ensemble comprising individual parametric models trained over single omics to address the computational challenges posed by the high dimensionality and variation in feature counts across the different yet integrated multi-omics datasets. the multi-view ensemble improves the prediction of case vs. control and finds the most success in flagging a larger consistent set of associated features when compared with chance models, which may eventually be used downstream in identifying a novel composite biomarker signature of T1D risk. the current work demonstrates the utility of supervised ML in exploring integrated parallel multi-omics data in the ongoing quest for early T1D biomarkers, reinforcing the hope for identifying novel composite biomarker signatures of T1D risk via ML and ultimately informing early treatment decisions in the face of the escalating global incidence of this debilitating disease.

Reduced-Reference Learning for Target Localization in Deep Brain Stimulation.

This work proposes a supervised machine learning method for target localization in deep brain stimulation (DBS). DBS is a recognized treatment for essential tremor. The effects of DBS significantly depend on the precise implantation of electrodes. Recent research on diffusion tensor imaging shows that the optimal target for essential tremor is related to the dentato-rubro-thalamic tract (DRTT), thus DRTT targeting has become a promising direction. The tractography-based targeting is more accurate than conventional ones, but still too complicated for clinical scenarios, where only structural magnetic resonance imaging (sMRI) data is available. In order to improve efficiency and utility, we consider target localization as a non-linear regression problem in a reduced-reference learning framework, and solve it with convolutional neural networks (CNNs). The proposed method is an efficient two-step framework, and consists of two image-based networks: one for classification and the other for localization. We model the basic workflow as an image retrieval process and define relevant performance metrics. Using DRTT as pseudo groundtruths, we show that individualized tractography-based optimal targets can be inferred from sMRI data with high accuracy. For two datasets of 280x220/272x227 (0.7/0.8 mm slice thickness) sMRI input, our model achieves an average posterior localization error of 2.3/1.2 mm, and a median of 1.7/1.02 mm. The proposed framework is a novel application of reduced-reference learning, and a first attempt to localize DRTT from sMRI. It significantly outperforms existing methods using 3D-CNN, anatomical and DRTT atlas, and may serve as a new baseline for general target localization problems.

Multi-instance learning based lung nodule system for assessment of CT quality after small-field-of-view reconstruction

Small-field-of-view reconstruction CT images (sFOV-CT) increase the pixel density across airway structures and reduce partial volume effects. Multi-instance learning (MIL) is proposed as a weakly supervised machine learning method, which can automatically assess the image quality. The aim of this study was to evaluate the disparities between conventional CT (c-CT) and sFOV-CT images using a lung nodule system based on MIL and assessments from radiologists. 112 patients who underwent chest CT were retrospectively enrolled in this study between July 2021 to March 2022. After undergoing c-CT examinations, sFOV-CT images with small-field-of-view were reconstructed. Two radiologists analyzed all c-CT and sFOV-CT images, including features such as location, nodule type, size, CT values, and shape signs. Then, an MIL-based lung nodule system objectively analyzed the c-CT (c-MIL) and sFOV-CT (sFOV-MIL) to explore their differences. The signal-to-noise ratio of lungs (SNR-lung) and contrast-to-noise ratio of nodules (CNR-nodule) were calculated to evaluate the quality of CT images from another perspective. The subjective evaluation by radiologists showed that feature of minimal CT value (p = 0.019) had statistical significance between c-CT and sFOV-CT. However, most features (all with p < 0.05), except for nodule type, location, volume, mean CT value, and vacuole sign (p = 0.056–1.000), had statistical differences between c-MIL and sFOV-MIL by MIL system. The SNR-lung between c-CT and sFOV-CT had no statistical significance, while the CNR-nodule showed statistical difference (p = 0.007), and the CNR of sFOV-CT was higher than that of c-CT. In detecting the difference between c-CT and sFOV-CT, features extracted by the MIL system had more statistical differences than those evaluated by radiologists. The image quality of those two CT images was different, and the CNR-nodule of sFOV-CT was higher than that of c-CT.