Traditional Chinese medicine (TCM) is an ancient medical system distinctive and effective in treating cancer, depression, coronavirus disease 2019 (COVID-19), and other diseases. However, the relatively abstract diagnostic methods of TCM lack objective measurement, and the complex mechanisms of action are difficult to comprehend, which hinders the application and internationalization of TCM. Recently, while breakthroughs have been made in utilizing methods such as network pharmacology and virtual screening for TCM research, the rise of machine learning (ML) has significantly enhanced their integration with TCM. This article introduces representative methodological cases in quality control, mechanism research, diagnosis, and treatment processes of TCM, revealing the potential applications of ML technology in TCM. Furthermore, the challenges faced by ML in TCM applications are summarized, and future directions are discussed.

Objective : To construct a molecular classification system for head and neck squamous cell carcinoma (HNSCC) utilizing hypoxia-related gene (HAG) expression profiles, and to comprehensively examine the clinicopathological significance and biological functions of the hypoxia gene stanniocalcin 2 (STC2) in HNSCC. Methods : Transcriptomic data and clinical information of 546 HNSCC samples were obtained from The Cancer Genome Atlas (TCGA) database, and based on the expression profiles of 200 HRGs, HNSCC was classified subclasses using non-negative matrix factorization (NMF). HNSCC was classified into three subclasses (C1, C2, and C3), and the molecular characteristics and prognostic differences of the subclasses were assessed by comparing the tumor mutation load, functional enrichment analysis, drug sensitivity, and clinical features among the subclasses. LASSO-Cox regression was used to screen prognosis-related genes and construct prognostic models. Using oral squamous cell carcinoma (OSCC)-related data in the TCGA database, we analyzed the expression differences of STC2 in OSCC and control samples, and detected the mRNA and protein expression of STC2 in oral squamous carcinoma samples using qRT-PCR and immunohistochemistry. We knocked down STC2 in CAL-27 cells and verified the knockdown efficiency by qRT-PCR and Western blot. CCK-8 assay and cell scratch assay were used to assess the effect of STC2 on cell proliferation and migration ability. Results: Based on HRGs expression profiles, HNSCC was categorized into three subclasses (C1, C2, and C3). Subclass C1 had moderate hypoxic activity and good prognosis; subclass C2 had the highest hypoxic activity, poor prognosis, and poor sensitivity to CTLA-4 inhibitors (P<0.05); subclass C3 had the lowest hypoxic activity and moderate prognosis, and STC2 belonged to subclass C3. The frequency of cyclin-dependent kinase inhibitor 2A (CDKN2A), phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA), and tumor protein p53 (TP 53) mutations was higher in HNSCC. C1 genomic gain and deletion burden were significantly higher than C3 subclass (P<0.05) and C2 genomic gain than C3 subclass (P<0.05). The C2 subclass was significantly enriched in hypoxia-associated pathways, such as glycine metabolism and base excision repair (P<0.05). The C1, C2, and C3 subclasses were significantly positively correlated in terms of sex (male) (Cramer’s V=0.15), radiation exposure (Cramer’s V=0.12), medication (Cramer’s V=0.18), and pathological grading (G1/G2) (Cramer’s V=0.25) (P<0.05). Nine prognosis-related genes were screened by LASSO-Cox regression, among which high expression of STC2 was positively correlated with poorer overall survival (OS) in HNSCC patients (P<0.01). Bioinformatics analysis showed that STC2 mRNA expression was higher in OSCC than in normal controls (P<0.05). qRT-PCR and immunohistochemistry confirmed that both mRNA and protein expression of STC2 were significantly upregulated in OSCC tissues and cells (P<0.01). In vitro experiments showed that STC2 expression was knocked down to approximately 80% in CAL-27 cells (P<0.001), and the STC2 knockdown group had a reduced value-added rate (P<0.001) and a reduced percentage of scratch closure (P<0.05) compared with the control group. Conclusion We successfully constructed a molecular typing system for HNSCC based on the expression profiles of HRGs and categorized HNSCC into three subclasses with significant prognostic differences, among which the C2 subclass had the highest hypoxic activity and the poorest prognosis. STC2 was highly expressed in HNSCC and suggested a poor prognosis, demonstrating that it may be a potential target for HNSCC treatment.

Objective To analyze the differences of brain activity between first-episode untreated depressive patients with and without suicidal ideation(SI),and its correlations with clinical characteristics.Methods A total of 40 major depressive disorder(MDD)patients with SI(MDD+SI group),40 patients without SI MDD(MDD+NSI group),and 40 healthy controls(HC)(HC group)were enrolled.The 17-item Hamilton depression scale(HAMD-17)and Beck scale for suicide ideation(BSI)were used to assess the severity of depression and SI,respectively.MRI data were collected.The values of fractional amplitude of low-frequency fluctuation(fALFF)were calculated.Results(1)Compared with the HC group,the MDD+NSI group showed decreases in the fALFF val-ues of the default network and attention network.The fALFF values of the attention network in the MDD+SI group showed decreases.Compared with the MDD+NSI group,the MDD+SI group showed decreases in the fALFF values of the attention network.(2)The fALFF values in the left middle frontal gyrus were negatively correlated with the total score of HAMD-17(r=-0.55;P<0.001)in the MDD+NSI group,while the fALFF values in the left middle frontal gyrus were negatively correlated with the total score of HAMD-17(r=-0.53;P<0.001)and the total score of BSI(r=-0.51;P<0.001)in the MDD+SI group.(3)The optimal critical value of fALFF value in left middle frontal gyrus for predicting SI occurrence in MDD patients was-0.039,area under the curve(AUC)was 0.76,sensitivity was 0.63,and specificity was 0.80.Conclusion The decreased local activity intensity in the left middle frontal gyrus of the brain might be the central mechanism for the occurrence of SI in MDD patients.In addition,the left middle frontal gyrus might have certain value in identifying SI and predicting the severity of SI.

Traditional Chinese medicine(TCM)is an ancient medical system distinctive and effective in treating cancer,depression,coronavirus disease 2019(COVID-19),and other diseases.However,the relatively abstract diagnostic methods of TCM lack objective measurement,and the complex mechanisms of action are difficult to comprehend,which hinders the application and internationalization of TCM.Recently,while breakthroughs have been made in utilizing methods such as network pharmacology and virtual screening for TCM research,the rise of machine learning(ML)has significantly enhanced their inte-gration with TCM.This article introduces representative methodological cases in quality control,mechanism research,diagnosis,and treatment processes of TCM,revealing the potential applications of ML technology in TCM.Furthermore,the challenges faced by ML in TCM applications are summarized,and future directions are discussed.

In July 2024,the Diagnosis Related Groups(DRG)2.0 is released based on the Notice from the National Healthcare Security Administration on Issuing the DRG 2.0 and Deepening the Relevant Work.Compared with DRG 1.1,version 2.0 was established based on a wider range of suggestions regarding the Adjacent Diagnosis Related Groups(ADRG),Major Comorbidity or Complication(MCC),and Comorbidity or Complication(CC)from various institutions.A list of disease diagnoses and surgical operations that are not used as grouping rules was compiled,and grouping efficacy was further improved by upgrading the algorithms for MCC and CC with the help of AI.Meanwhile,it is necessary to pay more attention to the number of cases of ADRG,the better methods to list the MCC/CC,the suggestions of various doctors and continuously standardize the data and update the grouping scheme of DRG.

Objective To analyze the differences of brain activity between first-episode untreated depressive patients with and without suicidal ideation(SI),and its correlations with clinical characteristics.Methods A total of 40 major depressive disorder(MDD)patients with SI(MDD+SI group),40 patients without SI MDD(MDD+NSI group),and 40 healthy controls(HC)(HC group)were enrolled.The 17-item Hamilton depression scale(HAMD-17)and Beck scale for suicide ideation(BSI)were used to assess the severity of depression and SI,respectively.MRI data were collected.The values of fractional amplitude of low-frequency fluctuation(fALFF)were calculated.Results(1)Compared with the HC group,the MDD+NSI group showed decreases in the fALFF val-ues of the default network and attention network.The fALFF values of the attention network in the MDD+SI group showed decreases.Compared with the MDD+NSI group,the MDD+SI group showed decreases in the fALFF values of the attention network.(2)The fALFF values in the left middle frontal gyrus were negatively correlated with the total score of HAMD-17(r=-0.55;P<0.001)in the MDD+NSI group,while the fALFF values in the left middle frontal gyrus were negatively correlated with the total score of HAMD-17(r=-0.53;P<0.001)and the total score of BSI(r=-0.51;P<0.001)in the MDD+SI group.(3)The optimal critical value of fALFF value in left middle frontal gyrus for predicting SI occurrence in MDD patients was-0.039,area under the curve(AUC)was 0.76,sensitivity was 0.63,and specificity was 0.80.Conclusion The decreased local activity intensity in the left middle frontal gyrus of the brain might be the central mechanism for the occurrence of SI in MDD patients.In addition,the left middle frontal gyrus might have certain value in identifying SI and predicting the severity of SI.

In July 2024,the Diagnosis Related Groups(DRG)2.0 is released based on the Notice from the National Healthcare Security Administration on Issuing the DRG 2.0 and Deepening the Relevant Work.Compared with DRG 1.1,version 2.0 was established based on a wider range of suggestions regarding the Adjacent Diagnosis Related Groups(ADRG),Major Comorbidity or Complication(MCC),and Comorbidity or Complication(CC)from various institutions.A list of disease diagnoses and surgical operations that are not used as grouping rules was compiled,and grouping efficacy was further improved by upgrading the algorithms for MCC and CC with the help of AI.Meanwhile,it is necessary to pay more attention to the number of cases of ADRG,the better methods to list the MCC/CC,the suggestions of various doctors and continuously standardize the data and update the grouping scheme of DRG.

Purpose To analyze the multimodal imaging manifestations of phosphaturic mesenchymal tumor.Materials and Methods A retrospective analysis was performed on ten patients with phosphaturic mesenchymal tumor confirmed by pathology and molecular imaging from November 2012 to June 2022 in Tongji Hospital,Tongji Medical College,Huazhong University of Science and Technology.The imaging features of CT in four cases,of MRI in seven cases,and of nuclear medicine in ten cases were reviewed,along with the clinical characteristics of all patients.Results The main clinical symptoms of ten patients,eight cases had pain,four cases had weakness,and nine cases had multiple fractures.Laboratory tests results:ten cases showed decreased blood phosphorus,six cases showed decreased blood 25-hydroxyvitamin D and ten cases showed elevated blood alkaline phosphatase.The lesions were heterogeneous in four cases on CT scans and in six cases on MRI scans.Calcification was common,and some had cysts and fatty components.After enhancement,seven cases showed moderate to obvious uneven enhancement.99Tcm-MDP SPECT showed increased uptake of multiple bones and joints in eight cases,18F-FDG PET/CT showed no or slight increase in lesion metabolism in eight cases,68Ga-DOTATATE PET/CT showed significantly high uptake at the lesion site in ten cases.Conclusion Phosphaturic mesenchymal tumor patients typically present with pain,fractures and hypophosphatemia.The conventional imaging features are characterized by small and concealed lesions and complex components.Particularly,68Ga-DOTATATE PET/CT imaging exhibits high sensitivity for the tumor detection.

Objective Evaluation of the efficacy of high-frequency repetitive transcranial magnetic stimulation for chronic disturbance of consciousness after severe craniocerebral injury based on magnetic resonance spectroscopy.Methods The clinical data of 106 patients with chronic disturbance of consciousness after severe craniocerebral injury from January 2018 to December 2022 were retrospectively analyzed,and they were divided into control group(conventional rehabilitation treatment)and observation group(high frequency repetitive transcranial magnetic stimulation treatment)by propensity score matching method(1∶1),with 53 cases in each group.Both groups were examined by magnetic resonance spectroscopy(MRS)before and after treatment.The brain metabolic indexes[N-acetyl aspartate(NAA)/creatine(Cr)value,choline complex(Cho)/Cr value],Glasgow coma scale(GCS)score,electroencephalogram(EEG)grading,coma recovery scale(CRS-R)score,brainstem auditory evoked potential(BAEP)grading,upper limb sensory evoked potential(SSEP)grading and Cerebral blood flow perfusion index[cerebral blood volume(CBV),mean transit time(MTT),cerebral blood flow(CBF)]were compared between the two groups.Results After treatment,the NAA/Cr values of the thalamus and brainstem in the two groups increased,while the Cho/Cr values decreased,and the levels of brain metabolic indexes in the observation group were signifi-cantly better than those in the control group(P<0.05).The two groups'GCS score and CRS-R score increased,and the improvement of the observation group was better than that of the control group(P<0.05).The BAEP grading,EEG grading,and SSEP grading of the two groups improved,and those of the observation group were better than the control group(P<0.05).The CBF and CBV of the two groups increased,and MTT decreased,and the level of cere-bral blood perfusion index in the observation group was better than that in the control group(P<0.05).Conclusion High frequency repetitive transcranial magnetic stimulation has a significant effect on the recovery of patients with chronic consciousness disorders after severe craniocerebral injury.The mechanism may be related to improving the blood flow velocity of brain tissue and metabolism in the brain.

Objective:To analyze the predictive value of magnetic resonance imaging (MRI) examination in secondary dementia of cognitive impairment of patients with Parkinson disease (PD). Methods:A total of 100 PD patients with cognitive dysfunction,who admitted to the Affiliated Suzhou Hospital of Nanjing University Medical School and the Second Affiliated Hospital of Soochow University,were selected. They were divided into observation group (30 cases) and control group (70 cases) according to the Hasegawa Dementia Scale (HDS) score. The HDS score of observation group was ≤ 21.5 points,which had occurred secondary dementia,while the HDS score of control group was>21.5 points,which had not occurred secondary dementia. The imaging parameters,which included apparent diffusion coefficient (ADC) and fractional anisotropy (FA) index of magnetic resonance imaging (MRI),between two groups were compared. Logistic regression analysis was used to identify the risk factors for secondary dementia of PD patients with cognitive impairment. Receiver operating characteristic (ROC) curve was used to analyze the predictive efficacy of MRI imaging parameters for the secondary dementia of PD patients with cognitive impairment. Results:Compared with substantia nigra of the control group,the observation group had higher ADC values and lower FA values in substantia nigra,and the differences of them between the two groups were statistically significant (t=2.837,4.581,P<0.05),respectively. The ADC value and FA value of substantia nigra,age,PD duration and homocysteine (Hcy) were risk factors for the secondary dementia of cognitive impairment of PD patients (OR=1.537,1.346,1.134,1.351,1.625,P<0.05),respectively. The area under curve (AUC) value,sensitivity,specificity of the combined detection of ADC value and FA value of substantia nigra were respectively 0.861,94.86％ and 82.18％ in predict the secondary dementia of cognitive impairment of PD patients,and the predictive efficiency of combined detection was higher (95％CI:0.812-0.960,P<0.05). Conclusion:The increase of ADC value and the decrease of FA value of substantia nigra closely relate to the secondary dementia of cognitive dysfunction of PD patients. The combined detection of ADC value and FA value of substantia nigra can improve predictive efficiency of dementia.