Purpose To investigate the feasibility of multi-sequence MRI-based radiomics for predicting epidermal growth factor receptor(EGFR)mutation status in brain metastases from non-small cell lung cancer(NSCLC).Materials and Methods This retrospective study included 237 patients with NSCLC brain metastases from the Second Xiangya Hospital of Central South University(January 1,2017 to December 31,2023)who underwent EGFR genetic testing.All patients underwent pretreatment brain MRI including contrast-enhanced T1-weighted,T2-weighted FLAIR,and T2-weighted sequences,along with chest CT for primary lung lesions.EGFR mutations were identified in 120 patients.Using December 31,2021 as the cutoff date,patients were divided into training(n=146)and validation(n=91)cohorts.Senior radiologists delineated brain metastases on multi-sequence MRI and primary lesions on CT.A total of 851 radiomic features were extracted using PyRadiomics.Following feature selection,machine learning models were constructed using support vector machine algorithm and compared with least absolute shrinkage and selection operator-derived radiomic signatures.Five models were developed:three single-sequence MRI models,a multi-sequence MRI fusion model,and a CT model,with diagnostic performance evaluated by area under the receiver operating characteristic curve.Results The multi-sequence MRI fusion model demonstrated superior performance across all imaging types.The least absolute shrinkage and selection operator and support vector machine models achieved training set area under the curve of 0.854(95%CI 0.748-0.960)and 0.948(95%CI 0.923-0.973),respectively,and validation set area under the curve of 0.810(95%CI 0.751-0.869)and 0.951(95%CI 0.917-0.985),respectively.The optimal prediction model utilized support vector machine algorithm with multi-sequence MRI features.Conclusion Pretreatment multi-sequence MRI radiomics combined with machine learning accurately predicts EGFR mutation status in NSCLC patients with brain metastases.

Purpose To investigate the feasibility of multi-sequence MRI-based radiomics for predicting epidermal growth factor receptor(EGFR)mutation status in brain metastases from non-small cell lung cancer(NSCLC).Materials and Methods This retrospective study included 237 patients with NSCLC brain metastases from the Second Xiangya Hospital of Central South University(January 1,2017 to December 31,2023)who underwent EGFR genetic testing.All patients underwent pretreatment brain MRI including contrast-enhanced T1-weighted,T2-weighted FLAIR,and T2-weighted sequences,along with chest CT for primary lung lesions.EGFR mutations were identified in 120 patients.Using December 31,2021 as the cutoff date,patients were divided into training(n=146)and validation(n=91)cohorts.Senior radiologists delineated brain metastases on multi-sequence MRI and primary lesions on CT.A total of 851 radiomic features were extracted using PyRadiomics.Following feature selection,machine learning models were constructed using support vector machine algorithm and compared with least absolute shrinkage and selection operator-derived radiomic signatures.Five models were developed:three single-sequence MRI models,a multi-sequence MRI fusion model,and a CT model,with diagnostic performance evaluated by area under the receiver operating characteristic curve.Results The multi-sequence MRI fusion model demonstrated superior performance across all imaging types.The least absolute shrinkage and selection operator and support vector machine models achieved training set area under the curve of 0.854(95%CI 0.748-0.960)and 0.948(95%CI 0.923-0.973),respectively,and validation set area under the curve of 0.810(95%CI 0.751-0.869)and 0.951(95%CI 0.917-0.985),respectively.The optimal prediction model utilized support vector machine algorithm with multi-sequence MRI features.Conclusion Pretreatment multi-sequence MRI radiomics combined with machine learning accurately predicts EGFR mutation status in NSCLC patients with brain metastases.

Objective: Tumor markers have been widely used clinically. Detection of serum CA125 is one of the commonly used clinical methods for early screening and early diagnosis of epithelial ovarian cancer, but it is difficult to diagnose epithelial ovarian cancer with a single specific tumor marker. In this study, the combinatorial tumor marker detection method was used to compare the value of each tumor marker alone and different combinations in the diagnosis of epithelial ovarian cancer. Methods: The clinical data of patients with epithelial ovarian cancer (n=65) and ovarian benign disease (n=29) were collected. Multiple tumor marker protein chip was used to detect cancer antigen 125 (CA125), carbohydrate antigen 242 (CA242), alpha-fetoprotein (AFP), beta-human chorionic gonadotropin (β-HCG), carcinoembryonic antigen (CEA), cancer antigen 199 (CA199), neuron-specific enolase (NSE), Ferritin, cancer antigen 153 (CA153), and human growth hormone (HGH) serum levels, and to compare the differences between the benign and malignant ovarian tumors. The correlation between tumor markers and clinicopathologic features for ovarian epithelial carcinoma was analyzed by χ2 test. Spearman rank analysis showed the correlation between CA125 expression level and other tumor markers in epithelial ovarian cancer and the correlation between age and the above 10 tumor markers. Sensitivity, specificity, positive predictive value, negative predictive value, Youden index, and diagnostic efficiency were used to evaluate the diagnostic value of single tumor marker and the combination of tumor markers. Results: The levels of β-HCG, NSE, CA153, and CA125 in the epithelial ovarian cancer group were higher than those in the ovarian benign disease group. The level of NSE in the serum of patients with epithelial ovarian cancer was related to the clinical stage of patients. In addition, the levels of CA242, β -HCG, CEA, NSE, Ferritin, CA153 in the serum of patients with epithelial ovarian cancer were positively correlated with CA125 (rs=0.497, P< 0.001; rs=0.612, P<0.001; rs=0.358, P=0.003; rs=0.680, P<0.001; rs=0.322, P=0.009; rs= 0.609, P<0.001, respectively), and the levels of β-HCG, Ferritin, CA153 were positively correlated with the patient's age (rs=0.256, P=0.040; rs=0.325, P=0.008; rs=0.249, P=0.046, respectively). In the diagnosis of epithelial ovarian cancer, the sensitivity, Youden index, and diagnostic efficiency of CA125 detection alone were higher than the results of the other 9 separate detections. When CA153, CA199, CA242, Ferritin, and CEA were combined with CA125, the sensitivity of the combined detection of different combinations was higher than that of CA125 alone. The combined detection sensitivities of CA125+CEA and CA125+Ferritin+CEA were 89.2％ and 90.8％, respectively, and the diagnostic efficiencies were both 84.1％, which were higher than those of other combinations. The Youden index of CA125+CEA joint detection was 0.616, which was higher than those of other combinations. Conclusion: CA125 has a high diagnostic value in the diagnosis of epithelial ovarian cancer. The detection of combined tumor markers in serum has higher sensitivity and specificity in epithelial ovarian cancer.

OBJECTIVES: To study the feasibility of ArcCHECK-3DVH system in dosimetric verification for stereotactic body radiaotherapy (SBRT) with flattening filter free (FFF) model. METHODS: SBRT treatment plans for 57 patients were introduced into ArcCHECK phantom and recalculated. The calculated dose distribution of treatment planning system and the measured dose distribution of ArcCHECK phantom were compared by γ analysis. Then the 3 dimensional dose distribution of target and organs at risk was reconstructed by 3DVH software. The reconstructed dose and calculated dose with treatment planning system (TPS) were compared, and the dose volume γ pass rate and deviation of dose volume parameters to the target and organs at risk were quantitatively valuated. RESULTS: Based on the threshold criteria (3%, 3 mm, 10%), namely the deviation of measuring points between the planned value and the measured value was less than 3%, and the proportion of points with similar values in the plane or sphere with the center of the point and the radius of 3 mm was 10%, the relative and absolute dose pass rates of SBRT treatment plans in ArcCHECK system via γ analysis were greater than 95%. Based on the stricter threshold criteria (2%, 2 mm, 10%), the relative and absolute dose pass rates of SBRT treatment plan in ArcCHECK system via γ analysis were about 93%. In 3DVH dose verification, the γ pass rate of target and organs at risk was exceed 97%, and the deviations in 3DVH of the target and organs at risk were less than ±5%. CONCLUSIONS The ArcCHECK-3DVH system in dose verification can provide more comprehensive dose distribution information to reasonably evaluate the SBRT plan, with more significance for guiding clinical treatment.
Humans ; Phantoms, Imaging ; Quality Assurance, Health Care ; Radiometry ; Radiosurgery ; Radiotherapy Dosage ; Radiotherapy Planning, Computer-Assisted ; Radiotherapy, Intensity-Modulated

To study the feasibility of ArcCheck verification system in dosimetric verification for stereotactic radiotherapy (SRT) the stereotactic radiotherapy (SRT) with flattening filter free (FFF) model.
 Methods: A total of 76 cases under SRT treatment plans were introduced into ArcCheck phantom and recalculated. Threshold criteria was set as (3%, 3 mm, 10%) or (2%, 2 mm, 10%). The calculated dose distribution and the measured dose distribution of ArcCheck phantom were compared by means of distance to agree (DTA) and Gamma analysis method respectively.
 Results: Based on the threshold criteria (3%, 3 mm, 10%), the relative and absolute mean pass rates of SRT treatment plans by DTA and Gamma analysis were greater than 95%. Based on the threshold criteria (2%, 2 mm, 10%), the relative and absolute mean pass rates of SRT treatment plan by DTA and Gamma analysis were about 90%. The dose pass rate of Gamma analysis method was slightly higher than that of DTA analysis method (P<0.001).
 Conclusion: The ArcCheck verification system is a rapid and accurate method for SRT dose verification, and discrepancies are found in different analysis methods.
Feasibility Studies ; Humans ; Phantoms, Imaging ; Radiosurgery ; methods ; Radiotherapy Dosage ; Radiotherapy Planning, Computer-Assisted ; Radiotherapy, Intensity-Modulated