Organoids derived from patients provide a new opportunity for research and individualized treatment of malignant peritoneal mesothelioma.

BACKGROUND: Malignant peritoneal mesothelioma (MPM) is an extremely rare and highly invasive tumor. Due to the lack of accurate models that reflect the biological characteristics of primary tumors, studying MPM remains challenging and is associated with an exceedingly unfavorable prognosis. This study was aimed to establish a new potential preclinical model for MPM using patient-derived MPM organoids (MPMOs) and to comprehensively evaluate the practicality of this model in medical research and its feasibility in guiding individualized patient treatment. METHODS: MPMOs were constructed using tumor tissue from MPM patients. Histopathological analysis and whole genome sequencing (WGS) were employed to determine the ability of MPMOs to replicate the original tumor's genetic and histological characteristics. The subcutaneous and orthotopic xenograft models were employed to assess the feasibility of establishing an in vivo model of MPM. MPMOs were also used to conduct drug screening and compare the results with retrospective analysis of patients after treatment, in order to evaluate the potential of MPMOs in predicting the effectiveness of drugs in MPM patients. RESULTS: We successfully established a culture method for human MPM organoids using tumor tissue from MPM patients and provided a comprehensive description of the necessary medium components for MPMOs. Pathological examination and WGS revealed that MPMOs accurately represented the histological characteristics and genomic heterogeneity of the original tumors. In terms of application, the success rate of creating subcutaneous and orthotopic xenograft models using MPMOs was 88% and 100% respectively. Drug sensitivity assays demonstrated that MPMOs have different medication responses, and these differences were compatible with the real situation of the patients. CONCLUSION: This study presents a method for generating human MPM organoids, which can serve as a valuable research tool and contribute to the advancement of MPM research. Additionally, these organoids can be utilized as a means to evaluate the effectiveness of drug treatments for MPM patients, offering a model for personalized treatment approaches.

Combined brain topological metrics with machine learning to distinguish essential tremor and tremor-dominant Parkinson's disease.

BACKGROUND: Essential tremor (ET) and Parkinson's disease (PD) are the two most prevalent movement disorders, sharing several overlapping tremor clinical features. Although growing evidence pointed out that changes in similar brain network nodes are associated with these two diseases, the brain network topological properties are still not very clear. OBJECTIVE: The combination of graph theory analysis with machine learning (ML) algorithms provides a promising way to reveal the topological pathogenesis in ET and tremor-dominant PD (tPD). METHODS: Topological metrics were extracted from Resting-state functional images of 86 ET patients, 86 tPD patients, and 86 age- and sex-matched healthy controls (HCs). Three steps were conducted to feature dimensionality reduction and four frequently used classifiers were adopted to discriminate ET, tPD, and HCs. RESULTS: A support vector machine classifier achieved the best classification performance of four classifiers for discriminating ET, tPD, and HCs with 89.0% mean accuracy (mACC) and was used for binary classification. Particularly, the binary classification performances among ET vs. tPD, ET vs. HCs, and tPD vs. HCs were with 94.2% mACC, 86.0% mACC, and 86.3% mACC, respectively. The most power discriminative features were mainly located in the default, frontal-parietal, cingulo-opercular, sensorimotor, and cerebellum networks. Correlation analysis results showed that 2 topological features negatively and 1 positively correlated with clinical characteristics. CONCLUSIONS: These results demonstrated that combining topological metrics with ML algorithms could not only achieve high classification accuracy for discrimination ET, tPD, and HCs but also help to reveal the potential brain topological network pathogenesis in ET and tPD.

Influence of vegetation pattern and aridity on soil properties related soil available water in the Mediterranean regions.

Grassland ecology is deteriorating along with a fall in biodiversity and ecosystem services as a result of climate change in the Mediterranean regions. Understanding the mechanism of feedback between soil properties related to available water and increasing aridity is a key component of preserving grassland ecosystems. Structural equation modelling was used to explore a deep understanding of the underlying mechanisms of the feedback between soil properties related to available water and increasing aridity. In most cases, vegetation patches had significantly higher soil properties related to available water than inter-patches. Compared to inter-patches, the fine fractions of silt and clay content, soil organic carbon, saturated hydraulic conductivity, and available water under vegetation patches increased by 3.79%-7.64%, 31.08%-37.64%, 96.65%-141.14%, and 2.63%-9.21%, respectively, under limestone and mica schist lithologies. The fine fractions of silt and clay content, soil organic carbon, and available water were more responsive to the aridity index than the vegetation patch, while saturated hydraulic conductivity was more responsive to the vegetation patch than the aridity index. These complex relationships demonstrated that the available water was significantly positively affected by the vegetation pattern (0.09) and the aridity index (0.21-0.38). Soil texture had a significantly direct effect (0.43-0.53) on available water. Increasing aridity would strengthen the contrast in soil water availability while weakening the contrast in saturated hydraulic conductivity between vegetation patches and inter-patches. Available water was controlled by many aspects except vegetation pattern and aridity index. Understanding these relationships helped in predicting and mitigating the impacts of climate change on soil properties related to available water. The study offered fresh perspectives on the mechanism of vegetation pattern and aridity index on the various soil properties related to available water in arid and semiarid grasslands ecosystems under climate change.

Cadmium influence on lipid metabolism in Sprague-Dawley rats through linoleic acid and glycerophospholipid metabolism pathways.

Cadmium (Cd) is widely distributed in the environment and easy adsorbed by living organisms with adverse effects. Exposure to Cd-contaminated food may disrupt lipid metabolism and increase human health risk. To study the perturbation effect of Cd on lipid metabolism in vivo, 24 male Sprague-Dawley (SD) rats were randomly assigned four groups and treated by Cd chloride solution (0, 1.375 mg/kg, 5.5 mg/kg, 22 mg/kg) for 14 days. The characteristic indexes of serum lipid metabolism were analyzed. Afterwards, untargeted metabolomics analysis was applied to explore the adverse effects of Cd on rats by liquid chromatography coupled with mass spectrometry (LC-MS). The results revealed that Cd exposure obviously decreased the average serum of triglycerides (TG) and low-density lipoprotein cholesterol (LDL-C) and caused an imbalance of endogenous compounds in the 22 mg/kg Cd-exposed group. Compared with the control group, 30 metabolites with significant differences were identified in the serum. Our results indicated that Cd caused lipid metabolic disorders in rats by disrupting linoleic acid and glycerophospholipid metabolism pathways. Furthermore, there were three kinds of remarkable differential metabolites-9Z,12Z-octadecadienoic acid, PC(20:4(8Z,11Z,14Z,17Z)/0:0), and PC(15:0/18:2(9Z,12Z)), which enriched the two significant metabolism pathways and could be the potential biomarkers.

Assessment of physiological posterior-tooth displacement under habitual occlusal force by intraoral scanning using implant-supported crowns as the reference.

STATEMENT OF PROBLEM: Studies that have used digital methods to quantitatively evaluate physiological tooth displacement under occlusal force are sparse. PURPOSE: The purpose of this clinical study was to measure physiological posterior tooth displacement under occlusal force by intraoral scanning and reverse engineering technology by using implants as the reference. MATERIAL AND METHODS: A total of 14 participants received 15 implant-supported single mandibular first molar crowns. The surface data of maxillary and mandibular posterior teeth (U1 and L1) and the buccal occlusal data in the maximum intercuspal position (MIP) with habitual occlusal force were obtained by using an intraoral scanner (TRIOS 3, v20.1.2). The U1 and L1 data were segmented into single teeth, which were then aligned to the buccal occlusal data by using the "best-fit alignment" command to build the data under occlusal force (U2 and L2). U1 and L1 data were compared with U2 and L2 data to calculate the centroid and functional cusp vertex displacements and the long axis deflections of the second premolars and second molars, taking the first molar as the reference. The medians, and first quartile (Q1), third quartile (Q3) of the above data were reported, and the Shapiro-Wilk and Wilcoxon tests were used to analyze the differences (α=.05). RESULTS: Under occlusal force, the median (Q1, Q3) centroid displacements of posterior teeth ranged from 61 (52, 101) µm to 146 (80, 186) µm; the functional cusp vertex displacements ranged from 82 (62, 117) µm to 146 (98, 189) µm, and the long axis deflections ranged from 0.45 (0.25, 0.87) degrees to 1.03 (0.52, 1.41) degrees. Mandibular second premolars displaced lingually, mesially, and apically; mandibular second molars displaced distally and apically; and maxillary second premolars and second molars displaced lingually and apically. CONCLUSIONS: A digital method taking implant-supported single crowns as the reference was used to demonstrate physiological posterior-tooth displacement under habitual occlusal force.

A Double Atomic-Tuned RuBi SAA/Bi@OG Nanostructure with Optimum Charge Redistribution for Efficient Hydrogen Evolution.

Charge redistribution on surface of Ru nanoparticle can significantly affect electrocatalytic HER activity. Herein, a double atomic-tuned RuBi SAA/Bi@OG nanostructure that features RuBi single-atom alloy nanoparticle supported by Bi-O single-site-doped graphene was successfully developed by one-step pyrolysis method. The alloyed Bi single atom and adjacent Bi-O single site in RuBi SAA/Bi@OG can synergistically manipulate electron transfer on Ru surface leading to optimum charge redistribution. Thus, the resulting RuBi SAA/Bi@OG exhibits superior alkaline HER activity. Its mass activity is up to 65000 mA mg-1 at an overpotential of 150 mV, which is 72.2 times as much as that of commercial Pt/C. DFT calculations reveal that the RuBi SAA/Bi@OG possesses the optimum charge redistribution, which is most beneficial to strengthen adsorption of water and weaken hydrogen-adsorption free energy in HER process. This double atomic-tuned strategy on surface charge redistribution of Ru nanoparticle opens a new way to develop highly efficient electrocatalysts.

Chlorine disinfection byproduct of diazepam affects nervous system function and possesses gender-related difference in zebrafish.

Chlorinated disinfection byproducts in water posed potential health threat to humans. Nowadays, chlorinated derivatives of diazepam were ubiquitously detected in drinking water. Among these derivatives, 2-methylamino-5-chlorobenzophenone (MACB) was capable of penetrating the blood-brain barrier (BBB) and induced microglial phagocytosis of neurons in zebrafish. However, little is known about the MACB metabolism in vivo. Here, we determined the metabolism of MACB in zebrafish and microglia cell model. We found that MACB mainly disrupted the metabolism of branched-chain amino acids (Leu, Ile and Val) in zebrafish model and gamma-aminobutyric acid (GABA) pathway-related amino acids in microglia model. Additionally, we demonstrated that MACB can be metabolized by the mixed-function oxidase CYP1A2 enzyme which could be inhibited by estrogen causing the gender-difference in the accumulation of MACB in vivo. These results indicated that MACB perturbed metabolism and induced neurological disorders, particularly in the female zebrafish.

Sodium dehydroacetate exposure decreases locomotor persistence and hypoxia tolerance in zebrafish.

Environmental exposure to sodium dehydroacetate (DHA-S) is inevitable as DHA-S is a high-volume preservative widely used in cosmetics, processed foods and personal care products. DHA-S is absorbed rapidly when administered orally or on the skin and generally considered to be safe and well tolerated. However, DHA-S has recently been reported to induce weight loss and allergic contact dermatitis, yet little is known about how DHA-S affect the related biological processes. Here, we characterize the biological effects of DHA-S on zebrafish model by directly waterborne exposure. Zebrafish is susceptible to DHA-S exposure at early developmental stage. DHA-S decreased the hatch rate and locomotor persistence of zebrafish, and eventually induced lethality during the continuous exposure at relatively low concentrations of commonly addition. Acute DHA-S exposure decreased respiration capacity in larval zebrafish, promoted the expression of HIF-1α (hypoxia-inducible factor-1α) and caused rapid adult zebrafish death in 30 h. We further demonstrated that DHA-S inhibited the activity of succinate dehydrogenase (SDH) inducing respiratory chain interruption, energy deficiency and organic acids accumulation. These results suggest that the approved DHA-S may pose serious environmental/ecological pressures on the aquatic animal's migration.

Chlorinated disinfection byproducts of diazepam perturb cell metabolism and induce behavioral toxicity in zebrafish larvae.

Numerous byproducts resulting from chlorinated disinfection are constantly being generated during water treatment processes. The potential risks of these new emerging pollutions remain largely unknown. Here, we determined the risks of chlorinated disinfection byproducts of diazepam (DZP) in the cellular and zebrafish exposure experiments. The cytotoxicity of disinfection byproducts (MACB and MBCC) was greater than DZP in macrophage raw 264.7 cells at 10 mg/L. We further found that the effects of MBCC on the metabolism of glycine, serine, threonine and riboflavin were far greater than DZP by the targeted metabolomics methods. Moreover, MBCC significantly decreased the peak amplitude of neuronal action potential in primary embryonic rat (Spragu-Dawley SD) hippocampal neurons. We finally determined behavioral toxicity of DZP and byproducts in zebrafish larvae. MBCC significantly decreased the maximal swim-activity and peak duration of zebrafish after 72 h exposure. Altogether, these findings indicate the MBCC pose serious pressures on public health.

Sodium dehydroacetate induces cardiovascular toxicity associated with Ca2+ imbalance in zebrafish.

The environmental effects of additives have attracted increasing attention. Sodium dehydroacetate (DHA-S), as an approved preservative, is widely added in processed foods, cosmetics and personal care products. However, DHA-S has been recently reported to induce hemorrhage and coagulation aberration in rats. Yet little is known about the ecotoxicological effect and underlying mechanisms of DHA-S. Here, we utilized the advantage of zebrafish model to evaluate such effects. DHA-S induced cerebral hemorrhage, mandibular dysplasia and pericardial edema in zebrafish after 24 h exposure (48-72 hpf) at 50 mg/L. We also observed the defective heart looping and apoptosis in DHA-S-treated zebrafish through o-dianisidine and acridine orange staining. Meanwhile, DHA-S induced the deficiency of Ca2+ and vitamin D3 in zebrafish. We further demonstrated that DHA-S stimulated Ca2+ influx resulting in Ca2+-dependent mitochondrial damage in cardiomyocytes. Additionally, DHA-S inhibited glucose uptake and repressed the biosynthesis of amino acids. Finally, we identified that sodium bicarbonate could rescue zebrafish from DHA-S induced cardiovascular toxicity. Altogether, our results suggest that DHA-S is a potential risk for cardiovascular system.

MoS2 @HKUST-1 Flower-Like Nanohybrids for Efficient Hydrogen Evolution Reactions.

A novel MoS2 -based flower-like nanohybrid for hydrogen evolution was fabricated through coating the Cu-containing metal-organic framework (HKUST-1) onto MoS2 nanosheets. It is the first time that MoS2 @HKUST-1 nanohybrids have been reported for the enhanced electrochemical performance of HER. The morphologies and components of the MoS2 @HKUST-1 flower-like nanohybrids were characterized by scanning electron microscopy, X-ray diffraction analysis and Fourier transform infrared spectroscopy. Compared with pure MoS2 , the MoS2 @HKUST-1 hybrids exhibit enhanced performance on hydrogen evolution reaction with an onset potential of -99 mV, a smaller Tafel slope of 69 mV dec-1 , and a Faradaic efficiency of nearly 100 %. The MoS2 @HKUST-1 flower-like nanohybrids exhibit excellent stability in acidic media. This design opens new possibilities to effectively synthesize non-noble metal catalysts with high performance for the hydrogen evolution reaction (HER).

A green method to synthesize flowerlike Fe(OH)3 microspheres for enhanced adsorption performance toward organic and heavy metal pollutants.

Dyestuffs and heavy metal ions in water are seriously harmful to the ecological environment and human health. Three-dimensional (3D) flowerlike Fe(OH)3 microspheres were synthesized through a green yet low-cost injection method, for the removal of organic dyes and heavy metal ions. The Fe(OH)3 microspheres were characterized by thermal gravimetric analysis (TGA), Fourier transform infrared (FT-IR), and transmission electron microscopy (TEM) techniques. The adsorption kinetics of Congo Red (CR) on Fe(OH)3 microspheres obeyed the pseudo-second-order model. Cr6+ and Pb2+ adsorption behaviors on Fe(OH)3 microspheres followed the Langmuir isotherm model. The maximum adsorption capacities of the synthesized Fe(OH)3 were 308, 52.94, and 75.64mg/g for CR, Cr6+, and Pb2+ respectively. The enhanced adsorption performance originated from its surface properties and large specific surface area of 250m2/g. The microspheres also have excellent adsorption stability and recyclability. Another merit of the Fe(OH)3 material is that it also acts as a Fenton-like catalyst. These twin functionalities (both as adsorbent and Fenton-like catalyst) give the synthesized Fe(OH)3 microspheres great potential in the field of water treatment.

Template-free synthesis of inorganic hollow spheres at water/"water-brother" interfaces as Fenton-like reagents for water treatment.

This paper reports a template-free method to synthesize a series of inorganic hollow spheres (IHSs) including Cu-1, Cu-2, Ni-1, Ni-2 based on mineralization reactions at water/"water-brother" interfaces. "Water-brother" was defined as a solvent which is miscible with water, such as ethanol and acetone. The water/"water-brother" interfaces are very different from water/oil interfaces. The "water-brother" solvent will usually form a homogenous phase with water. Interestingly, in our method, these interfaces can be formed, observed and utilized to synthesize hollow spheres. Utilizing the unique porous properties of the spheres, their potential application in water treatment was demonstrated by using Cu-1 IHSs as Fenton-like reagents for adsorption and decomposition of Congo Red from aqueous solution. The final adsorption equilibrium was achieved after 30min with the maximum adsorption capacity of 86.1mg/g, and 97.3% removal of the dye in 80min after adsorption equilibrium. The IHSs can be reused as least 5 times after treatment by NaOH. This method is facile and suitable for large-scale production, and shows great potential for watertreatment.

Evaluation on antithrombotic effect of aspirin eugenol ester from the view of platelet aggregation, hemorheology, TXB2/6-keto-PGF1α and blood biochemistry in rat model.

BACKGROUND: Based on the prodrug principle, aspirin and eugenol, as starting precursors, were esterified to synthesize aspirin eugenol ester (AEE). The aim of the present study was to evaluate the antithrombotic effect of AEE in an animal disease model. In order to compare the therapeutic effects of AEE and its precursors, aspirin, eugenol and a combination of aspirin and eugenol were designed at the same molar quantities as the AEE medium dose in the control group. METHODS: After oral administration of AEE (dosed at 18, 36 and 72 mg/kg) for seven days, rats were treated with k-carrageenan to induce tail thrombosis. Following the same method, aspirin (20 mg/kg), eugenol (18 mg/kg) and 0.5 % CMC-Na (30 mg/kg) were administered as control drug. Different drug effects on platelet aggregation, hemorheology, TXB2/6-keto-PGF1α ratio and blood biochemistry were studied. RESULTS: AEE significantly inhibited ADP and AA-induced platelet aggregation in vivo. AEE also significantly reduced blood and plasma viscosity. Moreover, AEE down-regulated TXB2 and up-regulated 6-keto-PGF1α, normalizing the TXB2/6-keto-PGF1α ratio and blood biochemical profile. In comparison with aspirin and eugenol, AEE produced more positive therapeutic effects than its precursors under the same molar quantity. CONCLUSION: It may be concluded that AEE was a good candidate for new antithrombotic and antiplatelet medicine. Additionally, this study may help to understand how AEE works on antithrombosis in different ways.

Lowering effects of aspirin eugenol ester on blood lipids in rats with high fat diet.

BACKGROUND: Aspirin and eugenol were esterified to synthesize aspirin eugenol ester (AEE). As a pale yellow and odourless crystal, AEE reduced the gastrointestinal damage of aspirin and vulnerability of eugenol. The study was conducted to evaluate the preventive effects of AEE on blood lipids in rats with high fat diet (HFD). METHODS: Suspensions of AEE and simvastatin were prepared in 5% carboxymethyl cellulose sodium (CMC-Na). In order to observe the intervention effects, the drugs and HFD were administrated at the same time. Based on individual weekly body weight (BW), AEE was intragastrically administrated at the dosage of 18, 36 and 54 mg/kg. Simvastatin (10 mg/kg) and CMC-Na (20 mg/kg) were used as control drug. After 6 weeks of administration, the changes of BW and blood lipid indices including triglyceride (TG), low density lipoprotein (LDL), high density lipoprotein (HDL) and total cholesterol (TCH) were determined in the experiment. RESULTS: The rat blood lipids profile in model group was remarkably different after feeding 6-weeks HFD. TG, TCH and LDL indexes in model group were increased significantly compared with those in control group (p < 0.01). AEE at the dosage of 54 mg/kg significantly decreased levels of TG, TCH and LDL (p < 0.01), and slowed the rate of BW gain in comparison with model group (p < 0.05). Moreover, high dose AEE showed better effects than simvastatin on reducing TCH level and similar effects on TG, HDL and LDL. CONCLUSION: AEE could remarkably reduce levels of TG, TCH and LDL in rats with high fat diet, and slow the rate of body weight gain. It was conducted that AEE was a potential candidate on reducing blood lipids level. The mechanism of action of AEE should be investigated in further studies.

Comprehensive assessment of the anti-obesity effects of highland barley total, insoluble, and soluble dietary fiber through multi-omics analysis.

The impact of different forms of dietary fiber (total, insoluble or soluble) derived from the same source on health remains incompletely understood. In this study, the effects of total, insoluble, and soluble dietary fiber extracted from highland barley (HDF, HIDF, and HSDF) on combating obesity were evaluated and compared. A high-fat diet (HFD) was used to induce obesity in a murine model, followed by gavage administration of HDF, HIDF, or HSDF, and a comprehensive multi-omics approach was utilized to assess and compare the effects of these dietary fibers on obesity-related parameters. The results showed that all three dietary fibers significantly reduced body weight, modified blood lipid profiles, and ameliorated tissue damage in HFD-fed mice. Additionally, 16S rRNA sequencing analysis of mice feces showed that three types of dietary fiber exerted varying degrees of impact on the composition and abundance of gut microbiota while simultaneously promoting the biosynthesis of short-chain fatty acids. Specifically, HDF supplementation remarkably enhanced the abundance of Coprococcus, while HIDF and HSDF supplementation elevated the levels of Akkermansia and Allobaculum, respectively. Transcriptomic and proteomic results suggested the PPAR signaling pathway as a central regulatory mechanism influenced by these fibers. HDF and HIDF were particularly effective in modulating biological processes related to triglyceride and fatty acid metabolism, identifying Abcc3 and Dapk1 as potential targets. Conversely, HSDF primarily affected processes related to membrane lipids, ceramides, and phospholipids metabolism, with Pck1 identified as a potential target. Collectively, HDF, HIDF, and HSDF demonstrated distinct mechanisms in exerting exceptional anti-obesity properties. These insights may inform the development of personalized dietary interventions for obesity.

Occurrence and health risk assessment of toxic metals and rare earth elements in microalgae: Insight into potential risk factors in new sustainable food resources.

Microalgae are a promising sustainable food source with high nutritional value and environmental benefits. This study investigated the presence of toxic metals and rare earth elements (REEs) in 68 microalgal-based food products and conducted a probabilistic risk assessment to evaluate potential health risks. The findings revealed high detection rates of REEs (80.96% to 100%) and heavy metals (83.82% to 100%), with REE concentrations ranging from 0.0055 to 0.5207 mg/kg. Heavy metals were detected at the following average concentrations: As (2.80 mg/kg) > Cr (1.27 mg/kg) > Pb (0.30 mg/kg) > Cd (0.20 mg/kg) > Hg (0.01 mg/kg). Carcinogenic risk analysis for Cd (3.004 × 10-3), Cr (1.484 × 10-3), and As (1.1283 × 10-2) indicated that 95th percentile values exceeded established safety thresholds (10-4). These findings highlight the critical need for stringent monitoring and the establishment of comprehensive regulatory frameworks for the safety of novel microalgae foods.

The role of bound polyphenols in the anti-obesity effects of defatted rice bran insoluble dietary fiber: An insight from multi-omics.

Considering the high abundance of bound polyphenols (BP) in whole grain dietary fiber (DF), this study utilized multi-omics approach to evaluate the impact of BP of defatted rice bran insoluble DF (RIDF) in modulating obesity. Mice on high-fat diet were gavage-administered RIDF, BP-removed or formulated RIDF. The results indicated that DF significantly reduced serum total cholesterol, triglycerides, high-density and low-density lipoprotein cholesterol levels. Moreover, hepatic lipid accumulation and damage induced by high-fat diet were significantly ameliorated with DF intervention. The presence of BP increased the abundance of beneficial bacteria g_Akkermansia and g_Butyricicocus, as well as the expression of butyric acid/propionic acid. Furthermore, the expression of hepatic lipids and lipid-like molecules was significantly decreased under the combined intervention of BP and DF, and this was accompanied by alterations in genes related to lipid, sterol, and cholesterol metabolic biological processes. These findings suggest that BP contribute to the anti-obesity effects of DF.

Machine learning models for diagnosis of essential tremor and dystonic tremor using grey matter morphological networks.

BACKGROUND: Essential tremor (ET) and dystonic tremor (DT) are the two most common tremor disorders, and misdiagnoses are very common due to similar tremor symptoms. In this study, we explore the structural network mechanisms of ET and DT using brain grey matter (GM) morphological networks and combine those with machine learning models. METHODS: 3D-T1 structural images of 75 ET patients, 71 DT patients, and 79 healthy controls (HCs) were acquired. We used voxel-based morphometry to obtain GM images and constructed GM morphological networks based on the Kullback-Leibler divergence-based similarity (KLS) method. We used the GM volumes, morphological relations, and global topological properties of GM-KLS morphological networks as input features. We employed three classifiers to perform the classification tasks. Moreover, we conducted correlation analysis between discriminative features and clinical characteristics. RESULTS: 16 morphological relations features and 1 global topological metric were identified as the discriminative features, and mainly involved the cerebello-thalamo-cortical circuits and the basal ganglia area. The Random Forest (RF) classifier achieved the best classification performance in the three-classification task, achieving a mean accuracy (mACC) of 78.7%, and was subsequently used for binary classification tasks. Specifically, the RF classifier demonstrated strong classification performance in distinguishing ET vs. HCs, ET vs. DT, and DT vs. HCs, with mACCs of 83.0 %, 95.2 %, and 89.3 %, respectively. Correlation analysis demonstrated that four discriminative features were significantly associated with the clinical characteristics. CONCLUSION: This study offers new insights into the structural network mechanisms of ET and DT. It demonstrates the effectiveness of combining GM-KLS morphological networks with machine learning models in distinguishing between ET, DT, and HCs.

Multitask-Guided Deep Clustering With Boundary Adaptation.

Multitask learning uses external knowledge to improve internal clustering and single-task learning. Existing multitask learning algorithms mostly use shallow-level correlation to aid judgment, and the boundary factors on high-dimensional datasets often lead algorithms to poor performance. The initial parameters of these algorithms cause the border samples to fall into a local optimal solution. In this study, a multitask-guided deep clustering (DC) with boundary adaptation (MTDC-BA) based on a convolutional neural network autoencoder (CNN-AE) is proposed. In the first stage, dubbed multitask pretraining (M-train), we construct an autoencoder (AE) named CNN-AE using the DenseNet-like structure, which performs deep feature extraction and stores captured multitask knowledge into model parameters. In the second phase, the parameters of the M-train are shared for CNN-AE, and clustering results are obtained by deep features, which is termed as single-task fitting (S-fit). To eliminate the boundary effect, we use data augmentation and improved self-paced learning to construct the boundary adaptation. We integrate boundary adaptors into the M-train and S-fit stages appropriately. The interpretability of MTDC-BA is accomplished by data transformation. The model relies on the principle that features become important as the reconfiguration loss decreases. Experiments on a series of typical datasets confirm the performance of the proposed MTDC-BA. Compared with other traditional clustering methods, including single-task DC algorithms and the latest multitask clustering algorithms, our MTDC-BA achieves better clustering performance with higher computational efficiency. Deep features clustering results demonstrate the stability of MTDC-BA by visualization and convergence verification. Through the visualization experiment, we explain and analyze the whole model data input and the middle characteristic layer. Further understanding of the principle of MTDC-BA. Through additional experiments, we know that the proposed MTDC-BA is efficient in the use of multitask knowledge. Finally, we carry out sensitivity experiments on the hyper-parameters to verify their optimal performance.