1.Study on The Anti-aging Effects of Longevity-enriched Metabolite Dimethylglycine
Jie HU ; Gong-Yu PU ; Jun-Lin LI ; Ju CAO ; Zhi-Xin LIN ; Wei-Wei AN ; Xue-Meng LI ; Jing AN
Progress in Biochemistry and Biophysics 2026;53(4):1048-1061
ObjectiveThe exacerbating trend of global population aging poses profound socioeconomic and public health challenges, making the comprehensive elucidation of biological aging mechanisms and the discovery of effective anti-aging interventions an urgent priority in the life sciences. Based on our previous serum metabolomics findings that dimethylglycine, an intermediate metabolite of amino acid metabolism naturally present in the human body, was significantly enriched in the serum of longevity families, this study aimed to systematically investigate the anti-aging effects of dimethylglycine both in living organisms and in controlled laboratory environments, and to preliminarily elucidate its underlying molecular mechanisms. While existing literature indicates that dimethylglycine possesses antioxidant and immunomodulatory properties, its direct anti-aging efficacy and the specific molecular pathways through which it operates remain largely unexplored. MethodsTo comprehensively evaluate the anti-aging properties of dimethylglycine, we utilized replicative senescent human embryonic lung fibroblasts, specifically the WI-38 cell line, as an experimental model in a controlled laboratory environment. Cell viability and safety were thoroughly assessed using Cell Counting Kit-8 and lactate dehydrogenase release assays across various concentrations of dimethylglycine. The impact of dimethylglycine on cellular senescence phenotypes, oxidative stress, and proliferative capacity was evaluated via senescence-associated beta-galactosidase staining, reactive oxygen species fluorescence detection, and 5-ethynyl-2'-deoxyuridine incorporation assays. Furthermore, the molecular alterations of senescence-associated secretory phenotype factors and core senescence signaling pathways were quantified using quantitative reverse transcription polymerase chain reaction for the messenger RNA levels of interleukin-6, interleukin-8, p21, and matrix metalloproteinase-1, and enzyme-linked immunosorbent assay for the measurement of p16 and p21 protein expression levels. For the living organism model, the wild-type nematode Caenorhabditis elegans was used to evaluate systemic physiological effects. We conducted a comprehensive lifespan analysis at 20°C, heat stress resistance survival assays at 35℃, senescence-associated beta-galactosidase staining, lipofuscin accumulation tracking, intracellular reactive oxygen species measurement, and Oil Red O staining to ascertain systemic lipid accumulation. Additionally, network pharmacology bioinformatics tools, including PharmMapper and STRING databases, and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis were utilized to predict target pathways, alongside highly detailed molecular docking simulations utilizing SwissDock and Protein-Ligand Interaction Profiler to examine interactions with the cytochrome P450 family 2 subfamily C member 9 protein. ResultsThe experimental outcomes robustly demonstrate the potent anti-aging capabilities of dimethylglycine. At the cellular level, toxicity analyses firmly confirmed that dimethylglycine is highly safe; continuous treatment with 50 mol/L and 70 mol/L of dimethylglycine for 5 d did not induce any cellular membrane damage or cytotoxicity, but rather actively promoted cellular proliferation. Utilizing the optimal standardized concentration of 50 mol/L, dimethylglycine treatment significantly ameliorated senescent phenotypic markers in human embryonic lung fibroblasts, which was evidenced by a drastic and highly significant reduction in the senescence-associated beta-galactosidase positive cell percentage (P<0.000 1) and intracellular reactive oxygen species levels (P<0.000 1), alongside a marked increase in the 5-ethynyl-2'-deoxyuridine-positive proliferation rate (P=0.003 5). On a molecular expression scale, dimethylglycine significantly downregulated the messenger RNA expression of multiple core senescence-associated secretory phenotype inflammatory factors, including interleukin-6, interleukin-8, p21, and matrix metalloproteinase-1. Concurrently, it effectively suppressed the protein expression of critical cell cycle arrest markers, diminishing p16 protein levels by 57.3% (P=0.000 4) and p21 protein levels by 27.2% (P=0.000 7). In the nematode Caenorhabditis elegans animal model, dimethylglycine significantly extended the mean lifespan from 20.402 d to an impressive 23.066 d (P<0.000 1) and notably enhanced overall survival rates under severe heat stress environmental conditions (P=0.017). Furthermore, systemic dimethylglycine intervention significantly mitigated age-related physiological decline by decreasing bodily lipofuscin accumulation (P<0.000 1), significantly reducing senescence-associated beta-galactosidase activity, lowering systemic reactive oxygen species fluorescence (P=0.008), and effectively alleviating overall fat accumulation (P<0.000 1). Mechanistically, extensive network pharmacology and Kyoto Encyclopedia of Genes and Genomes analyses strongly revealed that the potential targets of dimethylglycine are significantly enriched in fundamental drug metabolism and oxidative stress response pathways. Precision molecular docking simulations conclusively demonstrated that dimethylglycine forms highly stable structural interactions with the cytochrome P450 family 2 subfamily C member 9 protein, specifically highlighting the definitive formation of 5 stable hydrogen bonds involving serine 365, leucine 366, and serine 429 residues, as well as two critical salt bridge formations with arginine 97 and histidine 368 residues. It is additionally predicted to interact favorably with glutathione S-transferase family proteins. ConclusionDimethylglycine exhibits a profoundly significant and multifaceted anti-aging activity at both the cellular and entire living animal levels. By powerfully alleviating oxidative stress, heavily suppressing the core p16 and p21-dependent cellular senescence signaling pathways, and substantially mitigating the detrimental senescence-associated secretory phenotype, dimethylglycine effectively delays fundamental cellular senescence processes and drastically extends whole-organism lifespan. The biological mechanisms driving these robust protective effects are highly likely closely associated with its direct stable interactions with crucial metabolic and detoxifying enzyme systems, such as cytochrome P450 family 2 subfamily C member 9 and glutathione S-transferase family proteins, thereby systemically improving metabolic dysregulation and restoring critical redox homeostasis. This comprehensive study provides highly solid experimental evidence supporting dimethylglycine as a highly potent and safe potential anti-aging intervention agent, while simultaneously offering a clear molecular mechanistic explanation for the previously documented high abundance of dimethylglycine observed within exceptionally long-lived human populations.
2.Clinical Outcomes of Endoscopic Radiofrequency Stretta Therapy for Gastroesophageal Reflux Disease Treatment: A Retrospective Analysis From2 Tertiary Centers in Korea
Hyun LIM ; Yuri KIM ; Jin Hee NOH ; Jung In LEE ; Eun Jeong GONG ; Boram CHA ; Chan Hyuk PARK ; Da Hyun JUNG ; Ju Yup LEE ; Sun Hyung KANG ; In Kyung YOO ; Joo Young CHO ; Do Hoon KIM ;
Journal of Neurogastroenterology and Motility 2026;32(2):290-297
Background/Aims:
Endoscopic anti-reflux therapy is a therapeutic option for gastroesophageal reflux disease (GERD), providing durable effects. However, clinical data from Korea remain limited. This study evaluates the clinical outcomes of endoscopic radiofrequency Stretta therapy in Korean patients.
Methods:
A retrospective analysis was conducted on 71 patients with GERD who underwent Stretta therapy at 2 tertiary hospitals in Korea between November 2015 and July 2021. Clinical outcomes, including patient satisfaction, medication cessation or reduction, and complications, were evaluated. Pre- and post-procedural esophageal manometry and 24-hour pH monitoring test results were also analyzed.
Results:
Patient satisfaction rates at 1, 6, and 12 months post-procedure were 54.7% (35/64), 70.0% (28/40), and 75.0% (21/28), respectively. Medication cessation or reduction was achieved in 31.2% (20/64) at 1 month, 70.0% (28/40) at 6 months, and 67.9% (19/28) at 12 months. Esophageal manometry (n = 21) showed no significant changes in mean lower esophageal sphincter pressure (18.7 mmHg [2.5-52.9] vs 17.4 mmHg [0.0-43.0], P = 0.702) or mean integrated relaxation pressure (8.2 mmHg [0.0-28.0] vs 10.1 mmHg [0.0-31.0], P = 0.840). The 24-hour pH monitoring (n = 18) demonstrated a nonsignificant decrease in acid exposure time (pH < 4) from 2.3% (0.0-8.4) to 1.6% (0.0-7.3) (P = 0.182). Similarly, the DeMeester score decreased non-significantly from 8.4 (0.8-27.7) to 6.6 (0.8-21.8) (P = 0.352). No procedure-related complications occurred.
Conclusion
Endoscopic radiofrequency Stretta therapy appears to be a safe treatment option for GERD and may provide favorable patient satisfaction and medication reduction.
3.Prognostic Impact of Radiologic and Pathologic Features on the Development of Progressive Pulmonary Fibrosis in Patients With Interstitial Lung Disease Other Than Idiopathic Pulmonary Fibrosis
Hyeong Ryun CHO ; Myoung Ja CHUNG ; Hyemi CHOI ; Jinheum KIM ; Ae Ri AN ; Su Yeon AHN ; Jin Young YOO ; Gong Yong JIN ; David A LYNCH ; Kum Ju CHAE
Korean Journal of Radiology 2026;27(1):63-75
Objective:
To evaluate the prognostic impact of radiologic and pathologic features in patients with interstitial lung disease (ILD) other than idiopathic pulmonary fibrosis (IPF), and to identify the factors associated with the development of progressive pulmonary fibrosis (PPF) and survival.
Materials and Methods:
This study retrospectively enrolled 75 patients diagnosed with ILD other than IPF who underwent surgical lung biopsy between January 2004 and December 2020. Three chest radiologists independently reviewed the CT features and extent of fibrosis on preoperative and follow-up CT scans. Two pathologists reviewed the histopathological features, including the presence of interstitial pneumonia. The time to PPF and overall survival were estimated using the Kaplan-Meier method. The associations of CT and pathological features with PPF and all-cause mortality were examined using standard Cox regression and time-dependent Cox models, respectively.
Results:
A total of 75 non-IPF ILD patients (mean age ± standard deviations 56.4 ± 13.2 years; range, 40–88 years) were enrolled. The median follow-up duration was 75.3 months (range, 7.8–189.8 months). Traction bronchiectasis on CT (adjusted hazard ratio [HR], 6.40; P = 0.003) and body mass index (adjusted HR per 1-kg/m2 increase, 0.82; P = 0.002) were found to be significantly associated with PPF in multivariable analysis. Radiological progression (adjusted HR, 18.44;P < 0.001), symptomatic progression (adjusted HR, 4.19; P = 0.011), and age (adjusted HR for 1-year increase, 1.12; P < 0.001) were significantly associated with death.
Conclusion
Traction bronchiectasis on CT was a significant predictor of PPF, while radiologic and symptomatic progression and older age were associated with poorer survival in patients with ILD other than IPF. These findings indicate that careful radiological evaluation and symptom monitoring may help to predict disease progression and outcomes in patients with nonIPF ILD.
4.γ-Oryzanol Ameliorates Endothelial Replicative Senescence via Downregulation of SGLT2 Expression to Attenuate NADPH-Driven Oxidative Stress
Saugat SHIWAKOTI ; Kushal SHARMA ; Dal-Seong GONG ; Ju-Young KO ; In-Young LEE ; Hyun-Jung KIM ; Min-Ho OAK
Biomolecules & Therapeutics 2026;34(2):401-412
Replicative senescence in endothelial cells is characterized by an irreversible cell cycle arrest and impaired endothelial function, contributing to vascular aging and cardiovascular disease. Natural compounds are being actively studied for their potential to delay cellular senescence and protect vascular health. Among them, rice bran has demonstrated several vascular benefits that are mainly attributed to gamma-oryzanol (γ-Orz), a major bioactive component in rice bran. However, its role in regulating endothelial replicative senescence and the underlying molecular mechanisms remain unclear. This study aimed to explore the protective effects of rice bran extract (RBE) and γ-Orz on replicative senescence in porcine coronary artery endothelial cells (PCAECs). Replicative senescence was modeled in PCAECs by serial passaging from P1 to P3 with varying concentrations of RBE and γ-Orz.Senescence was evaluated by measuring senescence-associated β-galactosidase (SA-β-gal) activity, cell proliferation, oxidative stress, and the expression of cell cycle regulatory proteins. RBE and γ-Orz significantly reduced SA-β-gal activity, improved proliferation, and decreased oxidative stress in P3 cells, along with downregulation of senescence-related proteins p53, p21, and p16. Additionally, γ-Orz suppressed sodium-glucose co-transporter 2 expression, reduced NADPH oxidase overexpression, and restored eNOS levels. These findings indicate that RBE and γ-Orz delay endothelial senescence by alleviating oxidative stress, highlighting their potential to reduce cardiovascular disease risk associated with endothelial senescence.
5.Exploring the etiology,pathogenesis,and treatment methods of depression with somatization disorder from the perspective of"cold water,damp earth,and stagnant wood"
Zhihui LIU ; Hongxin ZHENG ; Baozhao JU ; Ye DONG ; Xue GONG ; Siyuan DING
Journal of Beijing University of Traditional Chinese Medicine 2025;48(10):1345-1351
Depression is primarily characterized by persistent low mood,emotional disorders,and loss of interest.Somatization disorder is a type of neurotic disorder,primarily featuring recurrent,varied,and non-organic physical discomfort symptoms.In traditional Chinese medicine,depression with somatization disorder can be classified under the category of"depression"and"hysteria"according to its clinical symptoms.This article focuses on the dynamic evolution of the functions of the kidney,spleen,and liver,and explains the core pathogenesis of depression with somatization disorder based on the theory of"cold water,damp earth,and stagnant wood"proposed by HUANG Yuanyu.It explores the causes and related manifestations.Symptoms mainly characterized by cramps and convulsions can be attributed to the liver;symptoms primarily characterized by heaviness,distension,and dull pain accompanied by weakness can be attributed to the spleen;symptoms primarily characterized by stiffness,reluctance to move or inability to move,and reluctance to extend outward can be attributed to the kidney.Liver depression and spleen dampness,as well as wood depression and earth congestion,are treated by strengthening the spleen and soothing the liver,drying dampness and calming wind,and the modified Xiaoyao Powder combined with Yueju Pill is recommended.Spleen dampness and kidney coldness,earth failure and water overflow,are treated by tonifying the spleen and kidney,warming water and transforming dampness,and the modified Jingui Shenqi Pill combined with Linggui Zhugan Decoction is recommended.Kidney coldness and liver hyperactivity,water not nourishing wood,are treated by warming the kidney and enriching essence,replenishing water and nourishing wood,and the modified Dihuang Yinzi is recommended.This article explores the corresponding etiology,pathogenesis,treatment methods,and prescriptions for depression with somatization disorder,with the aim of providing new insights for clinical diagnosis and treatment.
6.Ultrasound radiomics combined with machine learning for early diagnosis of seronegative hashimoto’s thyroiditis
Wenjun WU ; Chang LIU ; Shengsheng YAO ; Daming LIU ; Yuan LUO ; Yihan SUN ; Ting RUAN ; Mengyou LIU ; Li SHI ; Mingming XIAO ; Qi ZHANG ; Zhengshuai LIU ; Xingai JU ; Jiahao WANG ; Xiang FEI ; Li LU ; Yang GAO ; Ying ZHANG ; Liying GONG ; Xuanyu CHEN ; Wanli ZHENG ; Xiali NIU ; Xiao YANG ; Huimei CAO ; Shijie CHANG ; Zuoxin MA ; Jianchun CUI
Chinese Journal of Endocrine Surgery 2025;19(3):313-319
Objective:To evaluate the value of ultrasound radiomics combined with machine learning for early diagnosis of seronegative Hashimoto’s thyroiditis (SN-HT) .Methods:This retrospective study included 164 patients from Liaoning Provincial People’s Hospital , Lixin County People’s Hospital, Linghai Dalinghe Hospital, Fengcheng Phoenix Hospital, who underwent thyroidectomy for solitary nodules with normal thyroid function between Nov. 2016 and Jan. 2024. Postoperative pathology confirmed Hashimoto’s thyroiditis (HT) in some cases, who were further categorized into antibody-positive and antibody-negative groups based on serum antibody status. Patients without Hashimoto’s thyroiditis served as the control group. A total of 298 ultrasound images were analyzed. Radiomics features were extracted from hypoechoic non-nodular areas within 0.5 cm surrounding the tumor. Two senior pathologists and two senior ultrasound physicians independently assessed lymphocytic infiltration, eosinophilic changes of follicular epithelium, and the proportion of hypoechoic areas in pathology and ultrasound images, respectively. A machine learning model, CCH-NET, was developed using linear regression and t-distributed stochastic neighbor embedding (t-SNE) techniques. The dataset was divided into a training set (80%) and a validation set (20%) to compare the diagnostic accuracy of CCH-NET with that of senior ultrasound physicians. Results:In internal validation, CCH-NET achieved a diagnostic accuracy of 88.89% for both antibody-positive and antibody-negative groups, significantly higher than the 66.67% accuracy of senior ultrasound physicians ( P<0.01). In external validation, CCH-NET achieved 75.00% and 66.67% accuracy for the two groups, compared to 50.00% by senior ultrasound physicians. For the control group, both methods achieved 93.33% accuracy. The AUC of CCH-NET was 0.848, outperforming senior ultrasound physicians (0.681) ,demonstrating superior diagnostic performance. Conclusion:The radiomics-based CCH-NET model, using non-nodular hypoechoic areas as a specific indicator, can accurately identify early SN-HT in euthyroid patients. It significantly outperforms senior ultrasound physicians, improving diagnostic accuracy and reducing missed diagnoses.
7.Performance assessment of CyberKnife-based SBRT plans with VoLO and SO algorithm for liver cancer
Shaojuan WU ; Zhongjian JU ; Yu LI ; Hanshun GONG ; Baolin QU ; Xiaoliang LIU ; Shanshan GU ; Xiangkun DAI
China Medical Equipment 2025;22(6):7-13
Objective:To assess performance advantages of voxel-less optimization(VoLO)algorithm of CyberKnife-based S7 treatment plan system for the optimization of stereotactic body radiation therapy(SBRT)for liver cancer.Methods:The case data of 20 patients with hepatocellular carcinoma from Chinese PLA General Hospital during June 2022 and April 2023 were retrospectively selected,which included 10 patients with large hepatocellular carcinoma and 10 patients with small hepatocellular carcinoma.All patients adopted respectively sequential optimization(SO)and VoLO to conduct optimization for plan.The optimized quality of plan and execution efficiency of two kinds of algorithms were assessed,and the influences of different tumor volumes also were considered.The planed quality assessment included dosimetric parameters of the target region and organ at risk(OAR).The assessment parameters of execution efficiency included the numbers of monitor units(MUs),nodes and beams,and estimated treatment time.Paired t-test method was adopted to analyze quality of plan and treatment efficiency.Results:On the aspect of the dose of target region,for small hepatocellular carcinoma,the conformity index(CI)value(1.08±0.05)of target region of VoLO algorithm was significantly better than(1.17±0.06)of SO algorithm(t=4.631,P<0.05).The gradient index(GI),coverage rate and dose by 95%(D95%)of VoLO algorithm were better than those of SO algorithm,while the differences were not significant(P>0.05).According to the defined standards of liver surgery,for large hepatocellular carcinoma,the differences in CI,GI,coverage rate and D95%of target region between two kinds of algorithms were significant(t=3.337,4.238,-3.359,-3.311,P<0.05),respectively.On the aspect of dosimetry for OAR,for the target region of large hepatocellular carcinoma,the differences of liver Dmean and D700 cm3 between two kinds of algorithms were significant(t=4.114,3.415,P<0.05).However,for small hepatocellular carcinoma,there was no significant statistical difference in dosimetry parameters of OAR between two kinds of algorithms(P>0.05).The execution efficiency of the plan of VoLO group was obviously higher than that of SO group,and the differences of MU number,node number,beam number and estimated treatment time between two groups were significant(t=12.661,4.423,5.024,9.487,P<0.05).Conclusion:The quality of VoLO plan is significantly better than that of SO,which has a significant improvement in execution efficiency of treatment.For the cases of large hepatocellular carcinoma with more complexity,the VoLO optimization shows better advantages on the aspect of dose on target region,and protection for normal liver.
8.Performance assessment of CyberKnife-based SBRT plans with VoLO and SO algorithm for liver cancer
Shaojuan WU ; Zhongjian JU ; Yu LI ; Hanshun GONG ; Baolin QU ; Xiaoliang LIU ; Shanshan GU ; Xiangkun DAI
China Medical Equipment 2025;22(6):7-13
Objective:To assess performance advantages of voxel-less optimization(VoLO)algorithm of CyberKnife-based S7 treatment plan system for the optimization of stereotactic body radiation therapy(SBRT)for liver cancer.Methods:The case data of 20 patients with hepatocellular carcinoma from Chinese PLA General Hospital during June 2022 and April 2023 were retrospectively selected,which included 10 patients with large hepatocellular carcinoma and 10 patients with small hepatocellular carcinoma.All patients adopted respectively sequential optimization(SO)and VoLO to conduct optimization for plan.The optimized quality of plan and execution efficiency of two kinds of algorithms were assessed,and the influences of different tumor volumes also were considered.The planed quality assessment included dosimetric parameters of the target region and organ at risk(OAR).The assessment parameters of execution efficiency included the numbers of monitor units(MUs),nodes and beams,and estimated treatment time.Paired t-test method was adopted to analyze quality of plan and treatment efficiency.Results:On the aspect of the dose of target region,for small hepatocellular carcinoma,the conformity index(CI)value(1.08±0.05)of target region of VoLO algorithm was significantly better than(1.17±0.06)of SO algorithm(t=4.631,P<0.05).The gradient index(GI),coverage rate and dose by 95%(D95%)of VoLO algorithm were better than those of SO algorithm,while the differences were not significant(P>0.05).According to the defined standards of liver surgery,for large hepatocellular carcinoma,the differences in CI,GI,coverage rate and D95%of target region between two kinds of algorithms were significant(t=3.337,4.238,-3.359,-3.311,P<0.05),respectively.On the aspect of dosimetry for OAR,for the target region of large hepatocellular carcinoma,the differences of liver Dmean and D700 cm3 between two kinds of algorithms were significant(t=4.114,3.415,P<0.05).However,for small hepatocellular carcinoma,there was no significant statistical difference in dosimetry parameters of OAR between two kinds of algorithms(P>0.05).The execution efficiency of the plan of VoLO group was obviously higher than that of SO group,and the differences of MU number,node number,beam number and estimated treatment time between two groups were significant(t=12.661,4.423,5.024,9.487,P<0.05).Conclusion:The quality of VoLO plan is significantly better than that of SO,which has a significant improvement in execution efficiency of treatment.For the cases of large hepatocellular carcinoma with more complexity,the VoLO optimization shows better advantages on the aspect of dose on target region,and protection for normal liver.
9.Exploring the etiology,pathogenesis,and treatment methods of depression with somatization disorder from the perspective of"cold water,damp earth,and stagnant wood"
Zhihui LIU ; Hongxin ZHENG ; Baozhao JU ; Ye DONG ; Xue GONG ; Siyuan DING
Journal of Beijing University of Traditional Chinese Medicine 2025;48(10):1345-1351
Depression is primarily characterized by persistent low mood,emotional disorders,and loss of interest.Somatization disorder is a type of neurotic disorder,primarily featuring recurrent,varied,and non-organic physical discomfort symptoms.In traditional Chinese medicine,depression with somatization disorder can be classified under the category of"depression"and"hysteria"according to its clinical symptoms.This article focuses on the dynamic evolution of the functions of the kidney,spleen,and liver,and explains the core pathogenesis of depression with somatization disorder based on the theory of"cold water,damp earth,and stagnant wood"proposed by HUANG Yuanyu.It explores the causes and related manifestations.Symptoms mainly characterized by cramps and convulsions can be attributed to the liver;symptoms primarily characterized by heaviness,distension,and dull pain accompanied by weakness can be attributed to the spleen;symptoms primarily characterized by stiffness,reluctance to move or inability to move,and reluctance to extend outward can be attributed to the kidney.Liver depression and spleen dampness,as well as wood depression and earth congestion,are treated by strengthening the spleen and soothing the liver,drying dampness and calming wind,and the modified Xiaoyao Powder combined with Yueju Pill is recommended.Spleen dampness and kidney coldness,earth failure and water overflow,are treated by tonifying the spleen and kidney,warming water and transforming dampness,and the modified Jingui Shenqi Pill combined with Linggui Zhugan Decoction is recommended.Kidney coldness and liver hyperactivity,water not nourishing wood,are treated by warming the kidney and enriching essence,replenishing water and nourishing wood,and the modified Dihuang Yinzi is recommended.This article explores the corresponding etiology,pathogenesis,treatment methods,and prescriptions for depression with somatization disorder,with the aim of providing new insights for clinical diagnosis and treatment.
10.Ultrasound radiomics combined with machine learning for early diagnosis of seronegative hashimoto’s thyroiditis
Wenjun WU ; Chang LIU ; Shengsheng YAO ; Daming LIU ; Yuan LUO ; Yihan SUN ; Ting RUAN ; Mengyou LIU ; Li SHI ; Mingming XIAO ; Qi ZHANG ; Zhengshuai LIU ; Xingai JU ; Jiahao WANG ; Xiang FEI ; Li LU ; Yang GAO ; Ying ZHANG ; Liying GONG ; Xuanyu CHEN ; Wanli ZHENG ; Xiali NIU ; Xiao YANG ; Huimei CAO ; Shijie CHANG ; Zuoxin MA ; Jianchun CUI
Chinese Journal of Endocrine Surgery 2025;19(3):313-319
Objective:To evaluate the value of ultrasound radiomics combined with machine learning for early diagnosis of seronegative Hashimoto’s thyroiditis (SN-HT) .Methods:This retrospective study included 164 patients from Liaoning Provincial People’s Hospital , Lixin County People’s Hospital, Linghai Dalinghe Hospital, Fengcheng Phoenix Hospital, who underwent thyroidectomy for solitary nodules with normal thyroid function between Nov. 2016 and Jan. 2024. Postoperative pathology confirmed Hashimoto’s thyroiditis (HT) in some cases, who were further categorized into antibody-positive and antibody-negative groups based on serum antibody status. Patients without Hashimoto’s thyroiditis served as the control group. A total of 298 ultrasound images were analyzed. Radiomics features were extracted from hypoechoic non-nodular areas within 0.5 cm surrounding the tumor. Two senior pathologists and two senior ultrasound physicians independently assessed lymphocytic infiltration, eosinophilic changes of follicular epithelium, and the proportion of hypoechoic areas in pathology and ultrasound images, respectively. A machine learning model, CCH-NET, was developed using linear regression and t-distributed stochastic neighbor embedding (t-SNE) techniques. The dataset was divided into a training set (80%) and a validation set (20%) to compare the diagnostic accuracy of CCH-NET with that of senior ultrasound physicians. Results:In internal validation, CCH-NET achieved a diagnostic accuracy of 88.89% for both antibody-positive and antibody-negative groups, significantly higher than the 66.67% accuracy of senior ultrasound physicians ( P<0.01). In external validation, CCH-NET achieved 75.00% and 66.67% accuracy for the two groups, compared to 50.00% by senior ultrasound physicians. For the control group, both methods achieved 93.33% accuracy. The AUC of CCH-NET was 0.848, outperforming senior ultrasound physicians (0.681) ,demonstrating superior diagnostic performance. Conclusion:The radiomics-based CCH-NET model, using non-nodular hypoechoic areas as a specific indicator, can accurately identify early SN-HT in euthyroid patients. It significantly outperforms senior ultrasound physicians, improving diagnostic accuracy and reducing missed diagnoses.

Result Analysis
Print
Save
E-mail