Mixed Features Research Articles

Mixed Features and Nonfatal Suicide Attempt Among Individuals With Major Depressive Episode: Insights From the French MHGP Survey.

Mixed Features and Nonfatal Suicide Attempt Among Individuals With Major Depressive Episode: Insights From the French MHGP Survey.

CMMamba: channel mixing Mamba for time series forecasting

Transformer-based methods have achieved excellent results in the field of time series forecasting due to their powerful ability to model sequences and capture their long-term dependencies. However, Transformer-based methods are still computationally inefficient and have high costs when modeling long sequences. In addition, previous work ignored the destruction of the original in-channel features by too many mixed features when performing channel mixing. To address the above issues, we introduce Mamba into the field of time series forecasting. The selective state-space model Mamba models long sequences more efficiently by expanding linearly according to the sequence length. Based on this, we propose Channel Mixing Mamba (CMMamba), which uses bidirectional Mamba to model sequences and uses a channel mixing mechanism to select appropriate mixed features to enhance the feature representation within the original channel. By introducing Mamba, we achieve efficient modeling of sequences. We also propose Deep Convolutional Structure (DCS) to learn cross-channel dependencies and temporal order information. Extensive experimental results on six real-world public datasets demonstrate the effectiveness of CMMamba.

TFFD-Net: an effective two-stage mixed feature fusion and detail recovery dehazing network

TFFD-Net: an effective two-stage mixed feature fusion and detail recovery dehazing network

A closed-form continuous-depth neural-based hybrid difference features re-representation network for RUL prediction

Remaining Useful Life (RUL) prediction contributes to ensuring the reliability of mechanical systems and improving their maintenance plans. Recently, prediction methods based on deep learning have undergone rapid development. However, there are significant differences between multivariate monitoring sequences and utilizing a single model for feature extraction, which leads to reaching suboptimization. Additionally, focusing solely on a specific feature dimension usually imposes notable limitations on the model. This paper proposes a Closed-form Continuous-depth neural (CfC)-based hybrid difference Features Re-representation Network (CfC-F2RN) for RUL prediction. This method comprehensively utilizes hybrid difference features through two stages: initial feature representation and feature re-representation. In the initial feature representation stage, a Long Short-Term Memory (LSTM) is employed to capture the hidden state information of each time step in sequence data. Next, a novel attention mechanism is utilized to extract the hidden states and obtain a deep feature map. In the feature re-representation stage, the encoded features are re-represented using a decoder composed of the CfC model. Finally, the hidden state of the last LSTM unit in the encoder, serving as supplementary information, is combined with the decoded features to form a dual-latent feature representation of the mixed differential features. The prediction subnetwork is then adopted to accomplish RUL prediction. The superiority of CfC-F2RN is substantiated through benchmarking against established methods using the C-MAPSS dataset.

A Morphological Study of HLA-DR-Immunopositive Cells in Multiple Sclerosis Lesions and Their Implications for Pathogenesis

Background: Multiple sclerosis (MS) is characterized by white matter demyelinating plaques, which can be classified as active, chronic active, or chronic inactive based on the extent of demyelination, cellularity, and inflammation. Microglia and macrophages play a central role in these processes. This study aimed to investigate the morphological characteristics of HLA-DR-immunopositive cells in these plaques to improve our understanding of the roles of these cells in MS plaques. Methods: This study is a retrospective post-mortem histopathological study. We analyzed 90 plaques from 6 MS cases. Of the plaques studied, 77 were grouped into three categories: 28 active, 34 chronic active, and 15 chronic inactive. Additionally, five vacuolated white matter lesions, two axonal degeneration lesions, and six lesions with mixed histological features were included. Six control cases were also examined to assess HLA-DR-immunopositive cell expression across various age groups. The cells were classified based on their morphology into two types: round cells without processes (macrophages) and cells with varying processes and shapes (ramified microglia). Results: Both macrophages and ramified microglia were present in all lesion types, with a focus on identifying the predominant cell type. Of the 28 active plaques, macrophages were the primary cell type in 25 plaques, while ramified microglia predominated in 3. In the center of 49 chronic plaques, scattered ramified microglia were observed in 46, with three plaques showing a predominance of macrophages. Among the 34 chronic active lesions, ramified microglia were the main cell type in the periphery of 32 plaques, with the remaining two predominantly exhibiting macrophages. Conclusions: The predominance of macrophages in active lesions and the presence of scattered ramified microglia in the center of chronic plaques are consistent with the phagocytic role of macrophages. Meanwhile, the prevalence of ramified microglia at the periphery of chronic active lesions suggests a potential protective function in maintaining lesion stability.

A lighter hybrid feature fusion framework for polyp segmentation

Colonoscopy is widely recognized as the most effective method for the detection of colon polyps, which is crucial for early screening of colorectal cancer. Polyp identification and segmentation in colonoscopy images require specialized medical knowledge and are often labor-intensive and expensive. Deep learning provides an intelligent and efficient approach for polyp segmentation. However, the variability in polyp size and the heterogeneity of polyp boundaries and interiors pose challenges for accurate segmentation. Currently, Transformer-based methods have become a mainstream trend for polyp segmentation. However, these methods tend to overlook local details due to the inherent characteristics of Transformer, leading to inferior results. Moreover, the computational burden brought by self-attention mechanisms hinders the practical application of these models. To address these issues, we propose a novel CNN-Transformer hybrid model for polyp segmentation (CTHP). CTHP combines the strengths of CNN, which excels at modeling local information, and Transformer, which excels at modeling global semantics, to enhance segmentation accuracy. We transform the self-attention computation over the entire feature map into the width and height directions, significantly improving computational efficiency. Additionally, we design a new information propagation module and introduce additional positional bias coefficients during the attention computation process, which reduces the dispersal of information introduced by deep and mixed feature fusion in the Transformer. Extensive experimental results demonstrate that our proposed model achieves state-of-the-art performance on multiple benchmark datasets for polyp segmentation. Furthermore, cross-domain generalization experiments show that our model exhibits excellent generalization performance.

8126 PTH-Independent, Calcitriol-Mediated Hypercalcemia In Renal Cell Carcinoma

Abstract Disclosure: G.K. Rai: None. A. De Rosairo: None. K. Madani: None. D. Sacoto: None. A.A. Franco-Akel, MD: None. R. Belokovskaya: None. Hypercalcemia of malignancy is a known complication of advanced cancer. It occurs through various mechanisms, one of which is via 1,25-dihydroxyvitamin D (calcitriol). Calcitriol-mediated hypercalcemia accounts for approximately 1% of cases of hypercalcemia in the setting of malignancy and is usually seen in lymphomatous solid tumors. We present a gentleman with clear cell renal cell carcinoma associated with hypercalcemia with normal levels of intact parathyroid hormone (PTH), PTH-related protein (PTHrP), and 25-hydroxyvitamin D but elevated calcitriol levels. A 48 year-old-male with no significant past medical history with complaints of fatigue, hematuria, and weight loss of 38 pounds in 3 months was following urology for an elective right-sided nephrectomy. MRI of the abdomen and pelvis found a 16.1 cm renal mass with necrotic areas, causing mass effects on the liver, pancreas, and IVC. Preoperative laboratory showed hypercalcemia of 13.7 mg/dL (corrected 13.9 mg/dL) with microcytic anemia (hemoglobin 8.8 g/dL, MCV 72 fl). Workup showed low iPTH (3 pg/dL), 25-hydroxyvitamin D (16.9 ng/dL), and phosphorus level (1.7mg/dL) with elevated calcitriol (94.3 pg/dL) and a normal PTHrP. CT chest showed multiple pulmonary nodules on bilateral lobes, with the largest nodule ∼ 4.5 mm. Hypercalcemia was managed with IV fluids, calcitonin 300 units Q12 for two days, Denosumab IV infusion, prednisone 60 mg for three days, and oral ergocalciferol 50,000 units weekly. Calcium levels normalized, after which the patient underwent nephrectomy. Pathology revealed renal cell carcinoma grade G4 with mixed features of clear cell carcinoma with sarcomatoid and rhabdoid features. We report an incidental finding of hypercalcemia in a patient with renal mass. The elevated calcitriol levels can be attributed to ectopic calcitriol production through the renal mass. Post nephrectomy calcium levels remained within normal range for consecutive weeks. With this, we concur that hypercalcemia most likely originated from the renal mass that had ectopic production of calcitriol. Dysregulated calcitriol production is a well-known complication of lymphomas and granulomatous disease but is seldom reported with solid tumors. Usually, the etiology for hypercalcemia in patients with renal cell carcinoma is the overproduction of PTHrP, which was not the case with our patient. The ultimate resolution of hypercalcemia of malignancy was attained by treating the underlying malignancy. Presentation: 6/1/2024

8126 PTH-independent, Calcitriol-mediated Hypercalcemia in Renal Cell Carcinoma

Abstract Disclosure: G.K. Rai: None. A. De Rosairo: None. K. Madani: None. D. Sacoto: None. A.A. Franco-Akel, MD: None. R. Belokovskaya: None. Hypercalcemia of malignancy is a known complication of advanced cancer. It occurs through various mechanisms, one of which is via 1,25-dihydroxyvitamin D (calcitriol). Calcitriol-mediated hypercalcemia accounts for approximately 1% of cases of hypercalcemia in the setting of malignancy and is usually seen in lymphomatous solid tumors. We present a gentleman with clear cell renal cell carcinoma associated with hypercalcemia with normal levels of intact parathyroid hormone (PTH), PTH-related protein (PTHrP), and 25-hydroxyvitamin D but elevated calcitriol levels. A 48 year-old-male with no significant past medical history with complaints of fatigue, hematuria, and weight loss of 38 pounds in 3 months was following urology for an elective right-sided nephrectomy. MRI of the abdomen and pelvis found a 16.1 cm renal mass with necrotic areas, causing mass effects on the liver, pancreas, and IVC. Preoperative laboratory showed hypercalcemia of 13.7 mg/dL (corrected 13.9 mg/dL) with microcytic anemia (hemoglobin 8.8 g/dL, MCV 72 fl). Workup showed low iPTH (3 pg/dL), 25-hydroxyvitamin D (16.9 ng/dL), and phosphorus level (1.7mg/dL) with elevated calcitriol (94.3 pg/dL) and a normal PTHrP. CT chest showed multiple pulmonary nodules on bilateral lobes, with the largest nodule ∼ 4.5 mm. Hypercalcemia was managed with IV fluids, calcitonin 300 units Q12 for two days, Denosumab IV infusion, prednisone 60 mg for three days, and oral ergocalciferol 50,000 units weekly. Calcium levels normalized, after which the patient underwent nephrectomy. Pathology revealed renal cell carcinoma grade G4 with mixed features of clear cell carcinoma with sarcomatoid and rhabdoid features. We report an incidental finding of hypercalcemia in a patient with renal mass. The elevated calcitriol levels can be attributed to ectopic calcitriol production through the renal mass. Post nephrectomy calcium levels remained within normal range for consecutive weeks. With this, we concur that hypercalcemia most likely originated from the renal mass that had ectopic production of calcitriol. Dysregulated calcitriol production is a well-known complication of lymphomas and granulomatous disease but is seldom reported with solid tumors. Usually, the etiology for hypercalcemia in patients with renal cell carcinoma is the overproduction of PTHrP, which was not the case with our patient. The ultimate resolution of hypercalcemia of malignancy was attained by treating the underlying malignancy. Presentation: 6/1/2024

Texture-Based Classification of Fetal Growth Restriction From Intrauterine Neurosonographic Image.

Fetal growth restriction (FGR) is a condition where fetuses fail to reach their genetic potential for growth, posing a significant health challenge for newborns. The aim of this research was to explore the efficacy of texture-based analysis of neurosonographic images in identifying FGR in fetuses, which may provide a promising tool for early assessment of FGR. A retrospective analysis collected 100 intrauterine neurosonographic images from 50 FGR and 50 gestational age-appropriate fetuses. Using MaZda software, approximately 300 texture features were extracted from occipital white matter (OWM) and cerebellum of intrauterine neurosonographic images, respectively. Then 10 optimal features were separately selected by 3 algorithms, including the Fisher coefficient method, the method of minimizing classification error probability and average correlation coefficients, and the mutual information coefficient method. Further, the 10 statistically most significant features were selected from these sets to form the mixed feature set. After nonlinear discriminant analysis was performed to reduce feature dimensionality, the artificial neural network (ANN) classifier was conducted, respectively. For OWM and cerebellum, a total of 11 and 14 statistically significant features were selected. When the mixed feature sets of OWM and cerebellum were applied to ANN classifier, classification accuracy were 90.00% (κ = 0.800; P < .001) and 93.00% (κ = 0.860; P < .001), and the receiver operating characteristic curve for identifying FGR showed an area under the curve of 0.82 and 0.87. Texture analysis of fetal intrauterine neurosonographic images is a feasible and noninvasive strategy for evaluating FGR fetuses.

MA-MFCNet: Mixed Attention-Based Multi-Scale Feature Calibration Network for Image Dehazing

MA-MFCNet: Mixed Attention-Based Multi-Scale Feature Calibration Network for Image Dehazing

Trade-offs and thermodynamics of energy-relay proofreading.

Biological processes that are able to discriminate between different molecules consume energy and dissipate heat, using a mechanism known as proofreading. In this work, we thoroughly analyse the thermodynamic properties of one of the most important proofreading mechanisms, namely Hopfield's energy-relay proofreading. We discover several trade-off relations and scaling laws between several kinetic and thermodynamic observables. These trade-off relations are obtained both analytically and numerically through Pareto optimal fronts. We show that the scheme is able to operate in three distinct regimes: an energy-relay regime, a mixed relay-Michaelis-Menten (MM) regime and a Michaelis-Menten regime, depending on the kinetic and energetic parameters that tune transitions between states. The mixed regime features a dynamical phase transition in the error-entropy production Pareto trade-off, while the pure energy-relay regime contains a region where this type of proofreading energetically outperforms standard kinetic proofreading.

Enhancing soil nitrogen measurement via visible-near infrared spectroscopy: Integrating soil particle size distribution with long short-term memory models

Good quality of soil nitrogen data, which is essential for the advancement of both enhanced agricultural management and ecological environment, traditionally depends on labor intensive chemical procedures. Visible near-infrared (Vis-NIR) spectroscopy, acknowledged for its efficiency, environmental compatibility and rapidity, merges as a promising alternative. However, the effectiveness of Vis-NIR measurement models are significantly compromised by soil particle size distribution (PSD), presenting a substantial challenge in improving the measurement accuracy and reliability. Here an innovative deep learning methodology that integrates PSD with Vis-NIR spectroscopy was proposed for the measurement of nitrogen content in soil samples. By leveraging the LUCAS dataset, different strategies for integrating PSD with Vis-NIR spectral data in deep learning models were explored, revealing that our proposed InSGraL framework, which incorporated mixed features of PSD and spectra as LSTM inputs achieves superior performance. Compared to models utilizing solely Vis-NIR data, InSGraL exhibits a 39.47 % reduction in RMSE and a 42.55 % decrease in MAE, and demonstrates robust performance across various land cover types, achieving an R2 of 0.94 on grassland samples. Moreover, Shapley Additive exPlanations (SHAP) analysis revealed that incorporating PSD modifies the spectral input importance distribution, effectively mitigating spectral interference from particle size while highlighting critical wavelengths previously obscured. This study provides an innovative modeling strategy to mitigate the influence of PSD by integrating it within deep learning framework using Vis-NIR, contributing a deeper understanding of the relationship between PSD and Vis-NIR spectra for the measurement of nitrogen content and offering an effective means to attain soil nitrogen data.

Boosting Commit Classification Based on Multivariate Mixed Features and Heterogeneous Classifier Selection

Commit classification plays a crucial role in software maintenance, as it permits developers to make informed decisions regarding resource allocation and code review. There are several approaches for automatic commit classification, yet they do not sufficiently explore the features related to commits and consider the advantages of ensemble models over individual models. Therefore, there is some room for improvement. In this paper, we propose MuheCC, a commit classification approach based on multivariate mixed features and heterogeneous classifier selection to address these challenges. It mainly consists of three phases: (1) Multivariate mixed feature extraction, which extracts features from commit messages, changed code and handcrafted features to construct comprehensive mixed features; (2) Hyperparameter tuning based on genetic algorithm, which utilizes genetic algorithm to optimize candidate traditional models and ensemble models; (3) Heterogeneous classifier selection, which selects the optimal combinations of traditional and ensemble models, respectively, to build a heterogeneous classifier for commit classification. To evaluate this approach, we extend an existing dataset with code changes for each commit and compare MuheCC with three baselines on this real-world dataset. The results show that MuheCC outperforms all baselines, especially improving the best baseline by 7.25% for accuracy, 6.88% for precision, 7.25% for recall and 7.06% for [Formula: see text]-score. Furthermore, the ablation experiments validate that the performance advantage of MuheCC is mainly attributed to the multivariate features (e.g. 12.55% contributions to accuracy) and the heterogeneous classifier (e.g. 12.26% contributions to accuracy). We further discuss the impact of hyperparameter tuning and heterogeneous classifier selection on the performance of MuheCC. These results prove the superiority and potential practical value of MuheCC.

Reactive crystallization regulation for synthesizing NCM811 precursor by different impellers

The spherical and dense precursors with a narrow particle size distribution are essential to achieve high electrochemical performance in cathode materials. Four newly designed impellers, including the punched-flow turbine (PFT), punched-flow disc turbine (PFDT), curved-inclined-blade disc turbine (CIBDT), and 6-folded-pitched turbine (FPT-6), are applied to prepare the precursors to investigate the interplay among the fluid flow state, precursors’ morphology and growth, and physicochemical properties of cathodes. The flow and mixing features of double impellers are predicted using computational fluid dynamics, indicating that the FPT-6 plays a crucial role in crystal nucleation while PFDT facilitates crystal growth. This concurrence can be corroborated by examining the performance of precursors synthesized using these four impellers. It is first found that the electrochemical performance is distinctly enhanced by utilizing different impellers to regulate the specified nucleation and growth stage. By increasing the nucleation rate with double impellers of FPT-6, the tap density and sphericity of the precursor and cathode materials are significantly improved. The discharge capacity of cooperative agitators (FPT-6+PT) is 2.1 times that of the only propeller turbine at a high current density (10 C). The later growth stage is suitably regulated by FPT-6 or PFDT; properly extended reactive time can enhance the sphericity and electrochemical performance. The dual mixed-impeller (PT/CIBDT) combines intense mixing and shear ability, attaining a dense spheroid with a highly smooth surface. The cycle retention of this cathode is increased by 10.35% after 200 cycles at 1 C compared to a commercial counterpart. A regulatory strategy for the reactive co-precipitation process to enhance precursor stability is proposed here. Furthermore, the optimal operation parameters derived from this strategy can be effectively applied to commercial cathode materials.

SN 2021dbg: A Luminous Type IIP–IIL Supernova Exploding from a Massive Star with a Layered Shell

We present extensive observations and analysis of supernova (SN) SN 2021dbg, utilizing optical photometry and spectroscopy. For approximately 385 days following the explosion, SN 2021dbg exhibited remarkable luminosity, surpassing most Type II SNe (SNe II). This initial high luminosity is potentially attributed to interaction between the ejected material and the surrounding circumstellar material (CSM), as evidenced by the pronounced interaction signatures observed in its spectra. The subsequent high luminosity is primarily due to the significant 56Ni mass (0.17 ± 0.05 M ⊙) produced in the explosion. Based on the flux of flash emission lines detected in the initial spectra, we estimate that the CSM mass near the progenitor amounted to ∼(1.0–2.0) × 10−3 M ⊙, likely resulting from intense stellar wind activity 2–3 yr preceding the explosion. Considering the bolometric light curve, nebular spectrum modeling, and mass-loss rate, we suggest that the progenitor of SN 2021dbg was a red supergiant (RSG) with a mass of ∼20 M ⊙ and a radius of 1200 R ⊙. This RSG featured a thick hydrogen shell, which may have contained a region with a sharp decrease in material density, electron density, and temperature, contributing to its layered structure. This object demonstrates mixed features of Type IIP and IIL SNe, making it a transitional event linking the above two subclasses of SNe II.

Intratumoral heterogeneity of oncogenic drivers in mixed histology lung adenocarcinomas: How tissue selection impacts molecular testing?

Majority of the lung adenocarcinomas show a mixture of different histological patterns. The possibility of histologically heterogeneous areas of the adenocarcinoma showing genetic heterogeneity and harboring different driver mutations, with potentially significant clinical impact, has not been adequately addressed. Currently, there are no guidelines to suggest how to submit tumor tissue in adenocarcinomas with mixed histological features for molecular testing.The objective of this study is to assess intra-tumoral heterogeneity in prominent driver mutations among different morphological patterns of lung adenocarcinoma, its implications on the future of molecular testing as well as its potential impact on patient management.Twenty-three cases of mixed histology lung adenocarcinoma resected between 2018 and 2023 were retrieved from the archives. H&E slides were reviewed to identify the predominant and second most predominant histological patterns. The morphologically different tumor areas were manually macro-dissected for DNA extraction. Next-Generation Sequencing with Ion AmpliSeq™ Cancer Hotspot Panel v2 (Thermo Fisher Scientific, USA).Thirteen cases showed the same pathological variant in both histological components tested. Three cases (13 %) exhibited disparities in the variants detected across the different histological patterns tested (p=0.025). The discrepant findings had a direct therapeutic impact in 4.3 % cases. Seven cases showed no pathogenic variants detected on either of the histological components tested.This study elucidates the presence of infrequent yet significant intra-tumoral heterogeneity in the molecular profiles of mixed histology adenocarcinomas, highlighting the need for guidelines directing tissue selection for molecular testing to avoid missed therapeutic opportunities and mitigate disease relapse.

Prevalence and correlates of non-suicidal self-injury among patients with bipolar disorder: A multicenter study across China

BackgroundNon-suicidal self-injury (NSSI) is increasingly prevalent among patients with bipolar disorder (BD), raising concerns in psychology and mental health. Investigating the incidence and factors associated with NSSI is crucial for developing prevention and intervention strategies. MethodsNSSI behaviors were identified using the Ottawa Self-injury Inventory. The Clinically Useful Depression Outcome Scale supplemented with questions for the DSM-5 specifier of mixed features (CUDOS-M) and the Mini International Neuropsychiatric Interview (Hypo-)Manic Episode with Mixed Features-DSM-5 Module (MINI-M) were used to evaluate clinical symptoms. Non-parametric tests, chi-square tests, point-biserial correlation and logistic regression analyses were employed for the purposes of data analysis. ResultsThe enrolled sample comprised 1044 patients with BD from 20 research centers across China. Out of 1044 individuals, 446 exhibited NSSI behaviors, with 101 of them being adolescents, leading to a prevalence of 78.3 % among adolescent patients. The most common methods for females and males were “cutting” (41.2 %) and “hitting” (34.7 %), respectively. By binary logistic regression analysis, young age, female, bipolar type II disorder, with suicidal ideation and mixed states, depressive symptoms and without family history of mental disorder were correlates of NSSI in patients with BD (P < 0.05). LimitationsAs a cross-sectional study, causality between NSSI behaviors and associated factors cannot be established. Reporting and recall biases may occur due to self-rating scales and retrospective reports. ConclusionOur study indicates a concerning prevalence of NSSI, particularly among young patients with BD in China. Future research should focus on understanding NSSI behaviors in this population and developing effective interventions.

Thermodynamics of the reactive phase formation of the binder of the CoO–BaO–Al2O3 system for electromagnetic radiation protection

The article discusses the physicochemical basis for the production of barium aluminate binders, supplemented with compounds and solid solutions including cobalt oxide. The results of calculating the temperature dependence of Gibbs free energy are presented for model reactions simulating all possible variants of two-phase equilibria in the CoO–BaO–Al2O3 system. Using thermodynamic analysis, the stability of the terminals of the above system is established, which makes it possible to triangulate it. The complete subsolidus structure is presented, as well as the geometric and topological characteristics of the CoO–BaO–Al2O3 system and its phases. The presented results of thermodynamic analysis of solid-phase reactions involving barium aluminates and cobalt-containing oxides allow comparing the paths of reaction phase formation of binders from various ingredients. The article substantiates the possibility of effectively modifying barium-aluminum binders with waste cobalt catalysts on corundum supports. These circumstances make it possible to justify the choice of raw materials at a qualitative level, and, in combination with calculated data on the geometrostatistical characteristics of the subsolidus structure of the CoO–BaO–Al2O3 system, to obtain technologically important information about the features of dosage and mixing of components when fabricating the binder. The results of this study allow reducing the number of long-duration and labor-intensive experiments to obtain a modified binder with a target set of characteristics.

Thermodynamics of the reactive phase formation of the binder of the CoO–BaO–Al2O3 system for electromagnetic radiation protection

The article discusses the physicochemical basis for the production of barium aluminate binders, supplemented with compounds and solid solutions including cobalt oxide. The results of calculating the temperature dependence of Gibbs free energy are presented for model reactions simulating all possible variants of two-phase equilibria in the CoO–BaO–Al2O3 system. Using thermodynamic analysis, the stability of the terminals of the above system is established, which makes it possible to triangulate it. The complete subsolidus structure is presented, as well as the geometric and topological characteristics of the CoO–BaO–Al2O3 system and its phases. The presented results of thermodynamic analysis of solid-phase reactions involving barium aluminates and cobalt-containing oxides allow comparing the paths of reaction phase formation of binders from various ingredients. The article substantiates the possibility of effectively modifying barium-aluminum binders with waste cobalt catalysts on corundum supports. These circumstances make it possible to justify the choice of raw materials at a qualitative level, and, in combination with calculated data on the geometrostatistical characteristics of the subsolidus structure of the CoO–BaO–Al2O3 system, to obtain technologically important information about the features of dosage and mixing of components when fabricating the binder. The results of this study allow reducing the number of long-duration and labor-intensive experiments to obtain a modified binder with a target set of characteristics.

Combining Radiomics and Autoencoders to Distinguish Benign and Malignant Breast Tumors on US Images.

Background US is clinically established for breast imaging, but its diagnostic performance depends on operator experience. Computer-assisted (real-time) image analysis may help in overcoming this limitation. Purpose To develop precise real-time-capable US-based breast tumor categorization by combining classic radiomics and autoencoder-based features from automatically localized lesions. Materials and Methods A total of 1619 B-mode US images of breast tumors were retrospectively analyzed between April 2018 and January 2024. nnU-Net was trained for lesion segmentation. Features were extracted from tumor segments, bounding boxes, and whole images using either classic radiomics, autoencoder, or both. Feature selection was performed to generate radiomics signatures, which were used to train machine learning algorithms for tumor categorization. Models were evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity and were statistically compared with histopathologically or follow-up-confirmed diagnosis. Results The model was developed on 1191 (mean age, 61 years ± 14 [SD]) female patients and externally validated on 50 (mean age, 55 years ± 15]). The development data set was divided into two parts: testing and training lesion segmentation (419 and 179 examinations) and lesion categorization (503 and 90 examinations). nnU-Net demonstrated precision and reproducibility in lesion segmentation in test set of data set 1 (median Dice score [DS]: 0.90 [IQR, 0.84-0.93]; P = .01) and data set 2 (median DS: 0.89 [IQR, 0.80-0.92]; P = .001). The best model, trained with 23 mixed features from tumor bounding boxes, achieved an AUC of 0.90 (95% CI: 0.83, 0.97), sensitivity of 81% (46 of 57; 95% CI: 70, 91), and specificity of 87% (39 of 45; 95% CI: 77, 87). No evidence of difference was found between model and human readers (AUC = 0.90 [95% CI: 0.83, 0.97] vs 0.83 [95% CI: 0.76, 0.90]; P = .55 and 0.90 vs 0.82 [95% CI: 0.75, 0.90]; P = .45) in tumor classification or between model and histopathologically or follow-up-confirmed diagnosis (AUC = 0.90 [95% CI: 0.83, 0.97] vs 1.00 [95% CI: 1.00,1.00]; P = .10). Conclusion Precise real-time US-based breast tumor categorization was developed by mixing classic radiomics and autoencoder-based features from tumor bounding boxes. ClinicalTrials.gov identifier: NCT04976257 Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Bahl in this issue.