In recent years, with the rapid development of artificial intelligence technology, the application of deep learning in the field of pathology has been continuously expanding. Particularly, the rise of multimodal data fusion methods has opened up new technical paths for the precise diagnosis, prognosis assessment, and individualized treatment of tumors. By integrating multi-level and multi-source data such as clinical information, pathological omics, molecular omics, and imaging omics, deep learning models can identify potential associated features and key biological mechanisms that are difficult to reveal by a single modality, thereby significantly improving the accuracy of disease classification and the scientific nature of risk stratification. This article systematically reviews the research progress of multimodal data fusion methods based on deep learning in the field of pathology in recent years, focuses on sorting out different types of fusion strategies, evaluates their advantages and challenges in practical clinical applications, and looks forward to future development trends.

Histological staining is the basis of pathological analysis,but the traditional staining method relies on chemical reagents,which not only consumes a lot of resources,but also causes harm to the environment and human health.In recent years,with the rapid development of deep learning technology,virtual staining technology,as a new method,is expected to effectively replace and supplement the traditional histological staining methods.It uses neural networks to analyze unstained tissue images,generate digital images that are highly similar to chemical staining effects,and even realize the mutual conversion between different staining modes,reducing the laboratory's dependence on chemical reagents and providing sustainable research programs.In this paper,the basic principles of virtual staining and its potential applications in histopathology are introduced in detail,and the current challenges and future research directions are discussed.

[Objective]To study the clinical features of viral encephalitis with isolated dizziness,and to analyze the diagnostic efficacy of vestibular function examination and cerebrospinal fluid cytology in these patients.[Methods]Totally 37 cases of viral encephalitis with isolated dizziness and 10 healthy volunteers were included.Clinical data[dizziness handicap inventory(DHI)score,head imaging,electroencephalogram,vestibular function test,cerebrospinal fluid routine,biochemistry,cell morphology,etiology second-generation sequencing,misdiagnosis]were collected.The area under the ROC curve(AUC)of diagnostic value of each type of test was analyzed.The changes of each examination before and after treatment were compared.[Results]We found 89.19%(33/37)of the patients were misdiagnosed.Vestibular function smooth follow-up test indicated vestibular central lesion(AUC value:0.82)in 64.86%(24/37)of the patients.The number of CSF transformed lymphocytes increased in 86.49%(32/37)of the patients(AUC value:0.93),the CSF large lymphocytes increased in 97.30%(36/37)of the patients(AUC value:0.99),and the mononucleosis was activated in 94.59%(35/37)of the patients(AUC value:0.97).Furthermore,18.92%(7/37)of the patients had increased EEG slow wave(AUC value:0.60),while 13.51%(5/37)of the patients showed cortical swelling on head MR(AUC:0.60).After antiviral treatment,dizziness grade decreased(Z=-4.899,P<0.001),smooth tracking abnormalities decreased(Z=-4.583,P<0.001),the proportion of CSF transformed lymphocytes decreased(t=4.281,P<0.001),and the proportion of large lymphocytes decreased(t=6.905,P<0.001).[Conclusion]The misdiagnosis rate of viral encephalitis with isolated dizziness is high.Incorporating into diagnosis the increased large lymphocytes,transformed lymphocytes,activated monocytes in CSF cytology with smooth follow-up test may improve diagnostic efficiency.

Histological staining is the basis of pathological analysis,but the traditional staining method relies on chemical reagents,which not only consumes a lot of resources,but also causes harm to the environment and human health.In recent years,with the rapid development of deep learning technology,virtual staining technology,as a new method,is expected to effectively replace and supplement the traditional histological staining methods.It uses neural networks to analyze unstained tissue images,generate digital images that are highly similar to chemical staining effects,and even realize the mutual conversion between different staining modes,reducing the laboratory's dependence on chemical reagents and providing sustainable research programs.In this paper,the basic principles of virtual staining and its potential applications in histopathology are introduced in detail,and the current challenges and future research directions are discussed.

In recent years, with the rapid development of artificial intelligence technology, the application of deep learning in the field of pathology has been continuously expanding. Particularly, the rise of multimodal data fusion methods has opened up new technical paths for the precise diagnosis, prognosis assessment, and individualized treatment of tumors. By integrating multi-level and multi-source data such as clinical information, pathological omics, molecular omics, and imaging omics, deep learning models can identify potential associated features and key biological mechanisms that are difficult to reveal by a single modality, thereby significantly improving the accuracy of disease classification and the scientific nature of risk stratification. This article systematically reviews the research progress of multimodal data fusion methods based on deep learning in the field of pathology in recent years, focuses on sorting out different types of fusion strategies, evaluates their advantages and challenges in practical clinical applications, and looks forward to future development trends.

Objective:To explore the feasibility of staging Wilson's disease (WD) based on imaging indexes, and evaluate the clinical characteristic differences of WD patients at different stages.Methods:Sixty WD patients (40 with cerebral type and 20 with hepatic type) hospitalized in Department of Neurology, First Affiliated Hospital of Sun Yat-sen University from July 2015 to June 2022 and 20 age- and gender-matched normal controls were selected. All subjects accepted susceptibility-weighted imaging (SWI), diffusion tensor imaging (DTI), magnetic resonance spectroscopy (MRS); 7 regions of interest were selected: globus pallidus, caudate nucleus, putamen, thalamus, cerebellum, midbrain and pons; their fractional anisotropy (FA), corrected phase (CP), N-acetylaspartate/creatine (NAA/Cr) values were measured. According to DTI, SWI and MRS results, WD patients were divided into group of metal deposition stage (decreased CP, normal FA and NAA/Cr), group of fiber damage stage (abnormal FA, normal NAA/Cr), and group of neuron necrosis stage (decreased NAA/Cr); the clinical data (modified Young scale scores, Child-Pugh grading of liver function, serum copper content, and urinary copper content) and imaging indexes (FA, CP, and NAA/Cr) among the 3 groups and control group were compared. Results:Among the 60 patients, 19 patients were at metal deposition stage (including 18 with liver type and 1 with brain type), 28 patients at fiber injury stage (including 2 with liver type and 26 with brain type), and 13 patients at neuron necrosis stage (all brain type). (1) Compared with group of metal deposition stage and fiber damage stage, group of neuron necrosis stage had significantly decreased urinary copper content ( P<0.05). The modified Young scale scores in groups of metal deposition stage, fiber injury stage and neuronal necrosis stage increased successively; Child-Pugh grading in group of metal deposition stage was higher than that in groups of fiber injury stage and neuronal necrosis stage. (2) Compared with groups of metal deposition stage and neuron necrosis stage, CP values in the globus pallidus and substantia nigra in group of fiber injury stage group were significantly decreased ( P<0.05). Compared with groups of metal deposition stage and fiber injury stage, the FA value in the putamen and NAA/Cr value in the pallidum, thalamus and caudate nucleus in group of neuron necrosis stage were significantly decreased ( P<0.05). Conclusion:Disease stages of WD patients can be divided by imaging methods; neurological symptoms gradually worsen following progressed WD.

Objective To evaluate the clinical manifestations, metal metabolism, imaging characteristics and treatment response in patients with delayed Wilson disease (WD). Methods Patients with untreated WD (40 with delayed onset and 40 with non?delayed onset) were enrolled. Twenty healthy people were included as normal controls. All patients were evaluated with modified Young scale neural symptom scores, grade of Child liver function and mental symptoms rating scale, magnetic resonance imaging (MRI) scan, magnetic sensitive imaging (susceptibility weighted imaging, SWI), metal metabolism. Corrected phase (CP) was measured at SWI. After 2 week treatment, neurologic symptoms, liver function, and metal metabolism were reviewed. Results The total score of neurological symptoms in WD patients with delayed onset was lower than that of non?delayed onset (13.00 ± 6.87 vs. 21.13 ± 5.53, P=0.033). The scores of SCL?90 and HAMA depression scales in patients with delayed onset were lower than those of non?delayed onset. On T2 weighted imaging, areas including substantia nigra and thalamus, the caudate nucleus, globus pallidus, putamen presented high signal rate in patients with delated onset than those with non?delayed (P=0.022, 0.037, 0.022, 0.037, 0.029 respectively). The SWI CP values of cangbai sphere and shell nucleus in patients with delayed onset were lower than those with non?delayed onset. Patients with delayed onset had higher urinary copper than those with non?delayed onset before and after treatment (P=0.040, 0.036). After treatment, the score of abnormal tremor and gait in patients with delayed onset was decreased (P=0.037, 0.044), while as the occurrence of neurological symptoms was increased by 10%, and the liver function level in patients with delayed WD was decreased in 3 cases. Conclusions The brain of WD patients with delayed onset is mainly composed of metal deposits, however the cell damage is not apparent. Clinical symptoms are characterized by significant liver injury, but relatively mild neurological and psychiatric symptoms. Patients with delayed WD have higher urinary copper excretion than those with non?delayed WD. Chelating agents improves the neurological symptoms in patients with delayed onset.

The flower buds of Lonicera macranthoides (Shan Yin-Hua), represent an important traditional Chinese medicine and food ingredient. A phytochemical investigation of the 70% EtOH extract of the flower buds of L. macranthoides resulted in the isolation of 12 triterpenoids (1-12), including two new ursane-type nortriterpenes, 2α, 24-dihydroxy-23-nor-ursolic acid (1) and 2α, 4α-dihydroxy-23-nor-ursolic acid (2). Their structures were established by multiple spectroscopic methods and comparison with literature data. All isolated compounds were evaluated for their anti-inflammatory effects in LPS-activated RAW264.7 cells. Compounds 1 and 2 exhibited inhibitory effects on iNOS at the concentration of 30 μmol·L.
Animals ; Anti-Inflammatory Agents ; chemistry ; pharmacology ; Drugs, Chinese Herbal ; chemistry ; Enzyme Inhibitors ; chemistry ; pharmacology ; Ethanol ; chemistry ; Flowers ; chemistry ; Lonicera ; chemistry ; Macrophages ; drug effects ; metabolism ; Mice ; Molecular Structure ; Nitric Oxide ; metabolism ; Nitric Oxide Synthase Type II ; antagonists & inhibitors ; Plant Extracts ; chemistry ; Plants, Edible ; chemistry ; RAW 264.7 Cells ; Triterpenes ; chemistry ; pharmacology

Objective To investigate the value of neurite orientation dispersion and density imaging (NODDI) in quantitative evaluation of the microstructural changes in basal ganglia and thalamus in Wilson's disease (WD) patients,and to assess the diagnostic efficacy of NODDI.Methods Totally 27 WD patients (WD group) and 26 age-and sex-matched controls (control group) were enrolled.All subjects underwent MR scanning with NODDI.Parameters of NODDI,including intracellular volume fraction (Vic),orientation dispersion index (ODI) and isotropic volume fraction (Viso) of bilateral caudate nucleus,globus pallidus,putamen and thalamus were compared between the 2 groups.Correlation analysis was performed between each parameter of NODDI and clinical Young scores.Random Forest model was used to assess the relative importance of each parameter and to evaluate the diagnostic efficacy.Results The Vic and ODI of bilateral caudate nucleus,globus pallidus and putamen in WD patients were significantly lower than those in normal controls (all P＜0.05),while Viso was significantly higher than that in normal controls (all P＜0.05).The Vic of bilateral thalamus was lower,while Viso was higher in WD patients than those in normal controls (all P＜0.05),and ODI had no significant difference between the 2 groups (P=0.055).In WD patients,Vic and ODI of bilateral caudate nucleus,globus pallidus and putamen were negatively correlated with clinical Young scores.Viso of globus pallidus and putamen were positively correlated with clinical scores.The prediction accuracy of NODDI was 96.23％,and the area under ROC curve was 0.96.Conclusion NODDI can effectively evaluate changes in microstructures and metabolism during copper deposition in WD patients,and it may be useful in detecting changes of brain deep nuclei and assessing the progression of WD.

Objective To evaluate functional activity of the subcortical nuclei in Wilson's disease (WD) using resting state functional MRI (rs-fMRI),and to evaluate damage to the functional conjunction in the extracorticospinal tract in WD patients.Methods Twenty-two patients with WD (between January 2015 and January 2016),including 18 with cerebral type and 4 with hepatic type,and 20 age-matched healthy controls were enrolled.Neurological symptoms were scored using the modified Young Scale.Patients with cerebral type WD were divided into 4 subgroups.All study subjects underwent rs-fMRI of the brain.The values of amplitude of low-frequency fluctuation (ALFF) and regional homogeneity (REHO) in the thalamus,caudate nucleus,putamen and globus pallidus were determined.The relationships between rsfMRI metrics and clinical status were evaluated.Results ALFF values were lower in the caudate nucleus,putamen and right thalamus of WD patients than in controls (t =-3.07,-3.00,-3.12,-2.46,-2.20;P =0.005,0.006,0.004,0.020,0.036),while REHO values were lower in the left caudate nucleus and left thalamus of WD patients (t =-2.38,-2.16;P =0.025,0.040).In the caudate nucleus (P =0.032,0.029,0.023),thalamus (P =0.022,0.041,0.035) ALFF values were lower in group 4 than in other groups.REHO values of the putamen (P =0.040,0.017,0.040) and thalamus (P =0.024,0.029 7,0.041) were higher in group 4 than in other groups.ALFF values in the caudate nucleus (t =-0.29,P=0.037),and thalamus (t =-1.77,P =0.042) were lower,and REHO values in the caudate nncleus (t =-1.46,P =0.040) were lower,in patients of cerebral type than in hepatic type patients.Conclusions The damage to the functional activity of the subcortical nuclei may occur in the WD patients.The functional activity of nuclei may be different between hepatic and cerebral type patients.Damage to the activity of neurons in the putamen and thalamus may correlate with psychiatric symptoms in WD patients.