The existing epilepsy seizure detection algorithms have problems such as overfitting and poor generalization ability due to high reliance on manual labeling of electroencephalogram's data and data imbalance between seizure and interictal periods. An unsupervised learning detection method for epileptic seizure that jointed graph attention network (GAT) and Transformer framework (GAT-T) was proposed. In this method, channel correlations were adaptively learned by GAT encoder. Temporal information was captured by one-dimensional convolution decoder. Combining outputs of the two mentioned above, predicted values for electroencephalogram were generated. The collective anomaly score was calculated and the detection threshold was determined. The results demonstrated that GAT-T achieved the average performance exceeding 90% (or 99%) with a 0.25 s (or 2 s) time segment length, which could effectively detect epileptic seizures. Moreover, the channel association probability matrix was expected to assist clinicians in the initial screening of the epileptogenic zone, and ablation experiments also reflected the significance of each module in GAT-T. This study may assist clinicians in making more accurate diagnostic and therapeutic decisions for epilepsy patients.
Humans ; Electroencephalography/methods* ; Epilepsy/physiopathology* ; Algorithms ; Seizures/physiopathology* ; Neural Networks, Computer ; Signal Processing, Computer-Assisted

A novel epileptic seizure prediction prediction model based on multichannel temporal and spatial feature extractions is presented.The model applies Stockwell transform to the original multichannel electroencephalogram(EEG)signals for extracting time-frequency components.To address the issue of insignificant difference between preseizure and interseizure time-frequency features,an adaptive feature module composing of ConvNeXt,SENet and pyramid pooling module is designed to enhance the ability of capturing key time-frequency features in each EEG channel.Meanwhile,a prediction model based on Bi-NLSTM is constructed to improve the spatiotemporal dependence between the time-frequency features of multichannel high-order EEG for further promoting the epilepsy classification performance.On the CHB-MIT dataset,the model has an accuracy,sensitivity,specificity and AUC of 96.0%,97.8%,96.8%and 0.987,respectively,and the false positive rate per hour decreased to 0.038,outperforming the existing mainstream methods.In addition,the effect of each component on the model performance is verified by ablation study.The proposed method improves the overall performance for seizure prediction effectively by optimizing local time-frequency feature extraction and enhancing multichannel spatiotemporal dependence.

A novel epileptic seizure prediction prediction model based on multichannel temporal and spatial feature extractions is presented.The model applies Stockwell transform to the original multichannel electroencephalogram(EEG)signals for extracting time-frequency components.To address the issue of insignificant difference between preseizure and interseizure time-frequency features,an adaptive feature module composing of ConvNeXt,SENet and pyramid pooling module is designed to enhance the ability of capturing key time-frequency features in each EEG channel.Meanwhile,a prediction model based on Bi-NLSTM is constructed to improve the spatiotemporal dependence between the time-frequency features of multichannel high-order EEG for further promoting the epilepsy classification performance.On the CHB-MIT dataset,the model has an accuracy,sensitivity,specificity and AUC of 96.0%,97.8%,96.8%and 0.987,respectively,and the false positive rate per hour decreased to 0.038,outperforming the existing mainstream methods.In addition,the effect of each component on the model performance is verified by ablation study.The proposed method improves the overall performance for seizure prediction effectively by optimizing local time-frequency feature extraction and enhancing multichannel spatiotemporal dependence.

Objectives: This study aimed to present the Asia-Pacific consensus on long-term and sequential therapy for osteoporosis, offering evidence-based recommendations for the effective management of this chronic condition.The primary focus is on achieving optimal fracture prevention through a comprehensive, individualized approach. Methods: A panel of experts convened to develop consensus statements by synthesizing the current literature and leveraging clinical expertise. The review encompassed long-term anti-osteoporosis medication goals, first-line treatments for individuals at very high fracture risk, and the strategic integration of anabolic and anti resorptive agents in sequential therapy approaches. Results: The panelists reached a consensus on 12 statements. Key recommendations included advocating for anabolic agents as the first-line treatment for individuals at very high fracture risk and transitioning to anti resorptive agents following the completion of anabolic therapy. Anabolic therapy remains an option for in dividuals experiencing new fractures or persistent high fracture risk despite antiresorptive treatment. In cases of inadequate response, the consensus recommended considering a switch to more potent medications. The consensus also addressed the management of medication-related complications, proposing alternatives instead of discontinuation of treatment. Conclusions This consensus provides a comprehensive, cost-effective strategy for fracture prevention with an emphasis on shared decision-making and the incorporation of country-specific case management systems, such as fracture liaison services. It serves as a valuable guide for healthcare professionals in the Asia-Pacific region, contributing to the ongoing evolution of osteoporosis management.

Objective:To investigate the effect of single nucleotide polymorphism (SNP) of FCN gene on the susceptibility of pre-eclampsia (PE) in Han nationality pregnant women.Methods:A total of 274 PE pregnant women (PE group) and 154 healthy pregnant women (control group) admitted to Boai Hospital of Zhongshan, Affiliated Hospital to Southern Medical University from October 2020 to October 2022 were collected. The general information, medical history, reproductive history, blood pressure, body mass index and blood biochemical indicators before delivery were compared between the two groups. Twenty-three SNP loci of FCN gene family were genotyped by time-of-flight mass spectrometry, and the serum levels of ficolins (ficolin-1, -2 and -3) were detected by enzyme-linked immunosorbent assay.Results:(1) Compared with the control group, the body mass index, mean arterial pressure, gestational age at delivery, blood urea nitrogen, alanine aminotransferase, aspartate aminotransferase, direct bilirubin, albumin, and C-reactive protein in the PE group were significantly higher than those in the control group (all P<0.05). The levels of N-terminal pro-B type natriuretic peptide (NT-proBNP), placental growth factor (PlGF) and human soluble vascular endothelial growth factor receptor-1 (sFlt-1) were significantly different between the two groups (all P<0.05). (2) Among the 23 SNP loci in FCN gene family, 18 loci were in Hardy-Weinberg genetic equilibrium, including 5 loci in FCN1 gene, 10 loci in FCN2 gene, and 3 loci in FCN3 gene. Five loci that did not conform to Hardy-Weinberg genetic equilibrium were not included in the subsequent analysis. Compared with the control group, the genotype distribution of 3 loci of FCN2 gene (rs7872508, rs11103563, rs73664188) and 1 locus of FCN3 gene (rs3813800) in the PE group were significantly different (all P<0.05). After Bonferroni correction, only the genotype distribution of rs7872508 and rs73664188 in FCN2 gene were statistically different between the PE group and the control group (all P<0.05). Further analysis showed that for the rs7872508 locus of FCN2 gene, compared with GG genotype, genotype GT ( OR=3.025, 95% CI: 1.080-8.471) and TT ( OR=4.777, 95% CI: 1.758-12.979) both significantly increased the risk of PE (both P<0.05). For rs73664188 locus of FCN2 gene, compared with TT genotype, genotype TC ( OR=0.510, 95% CI: 0.334-0.778) significantly reduced the risk of PE ( P<0.05). (3) Compared with the control group, the serum levels of ficolin-1 and ficolin-2 in pregnant women in the PE group were significantly reduced (both P<0.05), while the level of ficolin-3 showed no significant change ( P=0.271). Correlation analysis showed that the serum levels of ficolin-2 in pregnant women in the PE group were significantly positively correlated with PlGF level ( r=0.321, P<0.001), and significantly negatively correlated with sFlt-1 level ( r=-0.187, P=0.002) and NT-proBNP level ( r=-0.392, P<0.001). Further analysis revealed that the serum levels of ficolin-2 in pregnant women of the PE group with GT and TT genotypes at rs7872508 locus of FCN2 gene were significantly reduced (both P<0.05), while the serum level of ficolin-2 in pregnant women of the PE group with TC genotype at the rs73664188 locus were significantly increased ( P<0.05). Conclusion:The SNP of FCN2 gene in FCN gene family might be related to the susceptibility to PE and have an effect on serum ficolin-2 level in PE pregnant women.

OBJECTIVE: To investigate the efficacy and safety of daratumumab in treatment of multiple myeloma (MM) patients with renal impairment (RI). METHODS: The clinical data of 15 MM patients with RI who received daratumumab-based regimen from January 2021 to March 2022 in three centers were retrospectively analyzed. Patients were treated with daratumumab or daratumumab combined with dexamethasone or daratumumab combined with bortezomib and dexamethasone and the curative effect and survival were analyzed. RESULTS: The median age of 15 patients was 64 (ranged 54-82) years old. Six patients were IgG-MM, 2 were IgA-MM,1 was IgD-MM and 6 were light chain MM. Median estinated glomerular filtration rate (eGFR) was 22.48 ml/(min·1.73 M₂). Overall response rate of 11 patients with MM was 91% (≥MR), including 1 case of stringent complete response (sCR), 2 cases of very good partial response (VGPR), 3 cases of partial response (PR) and 4 cases of minor response (MR). The rate of renal response was 60%(9/15), including 4 cases of complete response (CR), 1 case of PR and 4 cases of MR. A median time of optimal renal response was 21 (ranged 7-56) days. With a median follow-up of 3 months, the median progression-free survival and overall survival of all patients were not reached. After treatment with daratumumab-based regimen, grade 1-2 neutropenia was the most common hematological adverse reaction. Non-hematological adverse reactions were mainly infusion-related adverse reactions and infections. CONCLUSION Daratumumab-based regimens have good short-term efficacy and safety in the treatment of multiple myeloma patients with renal impairment.
Humans ; Middle Aged ; Aged ; Aged, 80 and over ; Multiple Myeloma/drug therapy* ; Retrospective Studies ; Dexamethasone/therapeutic use* ; Antibodies, Monoclonal/therapeutic use* ; Bortezomib/therapeutic use* ; Renal Insufficiency/drug therapy* ; Antineoplastic Combined Chemotherapy Protocols/therapeutic use*

Long non-coding RNA (lncRNA) is a hot topic in the field of researching tumor pathogenesis, and the importance in hematologic malignancies has been gradually being elucidated. LncRNA not only regulates hematological tumorigenesis and progression through affecting various biological processes such as cell proliferation, differentiation, pluripotency and apoptosis; moreover, abnormal expression and mutation of lncRNA are closely related to drug resistance and prognosis. Thus lncRNA can be used as novel biomarker and potential therapeutic target for hematological tumors. In this review, we will focus on the latest progress of lncRNA in hematological tumors to provide new ideas for the clinical diagnosis, prognostic evaluation together with research and development of target drugs for hematologic malignancies.
Humans ; RNA, Long Noncoding/metabolism* ; Hematologic Neoplasms/genetics* ; Neoplasms ; Carcinogenesis/pathology* ; Cell Transformation, Neoplastic/genetics* ; Gene Expression Regulation, Neoplastic

Humans ; Non-alcoholic Fatty Liver Disease ; Prospective Studies ; Risk Factors ; Exercise ; Liver

OBJECTIVE: To analyze the characteristics and prognosis of acute leukemia(AL) with SET-NUP214 fusion gene. METHODS: The clinical data of 17 patients over 14 years old newly diagnosed with SET-NUP214 positive AL admitted in Institute of Hematology and Blood Diseases Hospital from August 2017 to May 2021 were analyzed retrospectively. RESULTS: Among the 17 SET-NUP214 positive patients, 13 cases were diagnosed as T-ALL (ETP 3 cases, Pro-T-ALL 6 cases, Pre-T-ALL 3 cases, Medullary-T-ALL 1 case), AML 3 cases (2 cases M5, 1 case M0) and ALAL 1 case. Thirteen patients presented extramedullary infiltration at initial diagnosis. All 17 patients received treatment, and a total of 16 cases achieved complete remission (CR), including 12 cases in patients with T-ALL. The total median OS and RFS time were 23 (3-50) months and 21 (0-48) months, respectively. Eleven patients received allogeneic hematopoietic stem cell transplantation(allo-HSCT), with median OS time of 37.5 (5-50) months and median RFS time of 29.5 (5-48) months. The median OS time of 6 patients in chemotherapy-only group was 10.5 (3-41) months, and median RFS time of 6.5 (3-39) months. The OS and RFS of patients with transplantation group were better than those of chemotherapy-only group (P=0.038). Among the 4 patients who relapsed or refractory after allo-HSCT, the SET-NUP214 fusion gene did not turn negative before transplantation. While, in the group of 7 patients who have not relapsed after allo-HSCT till now, the SET-NUP214 fusion gene expression of 5 patients turned negative before transplantation and other 2 of them were still positive. CONCLUSION The fusion site of SET-NUP214 fusion gene is relatively fixed in AL patients, often accompanied by extramedullary infiltration. The chemotherapy effect of this disease is poor, and allo-HSCT may improve its prognosis.
Humans ; Adolescent ; Precursor T-Cell Lymphoblastic Leukemia-Lymphoma ; Retrospective Studies ; Leukemia, Myeloid, Acute/therapy* ; Hematopoietic Stem Cell Transplantation ; Acute Disease ; Prognosis ; Leukemia-Lymphoma, Adult T-Cell/therapy* ; Nuclear Pore Complex Proteins

OBJECTIVE: To investigate the clinical characteristics, diagnosis, and treatment of one patient with primary adrenal natural killer/T-cell lymphoma (PANKTCL), and to strengthen the understanding of this rare type of lymphoma. METHODS: The clinical manifestations, diagnosis and treatment process, and prognosis of the patient admitted in our hospital were retrospectively analyzed. RESULTS: Combined with pathology, imaging, bone marrow examination， etc, the patient was diagnosed with PANKTCL (CA stage, stage II; PINK-E score 3, high-risk group). Six cycles of "P-GemOx+VP-16" regimen（gemcitabine 1 g/m² d1 + oxaliplatin 100 mg/m² d 1 + etoposide 60 mg/m² d 2-4 + polyethylene glycol conjugated asparaginase 3 750 IU d 5） was performed, and complete response was assessed in 4 cycles. Maintenance therapy with sintilimab was administered after the completion of chemotherapy. Eight months after the complete response, the patient experienced disease recurrence and underwent a total of four courses of chemotherapy, during which hemophagocytic syndrome occurred. The patient died of disease progression 1 month later. CONCLUSION PANKTCL is rare, relapses easily, and has a worse prognosis. The choice of the "P-GemOx+VP-16" regimen combined with sintilimab help to improve the survival prognosis of patient with non-upper aerodigestive tract natural killer /T-cell lymphoma.
Humans ; Treatment Outcome ; Antineoplastic Combined Chemotherapy Protocols/therapeutic use* ; Retrospective Studies ; Etoposide ; Neoplasm Recurrence, Local/drug therapy* ; Asparaginase ; Deoxycytidine ; Lymphoma, T-Cell, Peripheral/drug therapy* ; Lymphoma, Extranodal NK-T-Cell/therapy* ; Oxaliplatin/therapeutic use*