Habitat Radiomics Based on MRI for Predicting Platinum Resistance in Patients with High-Grade Serous Ovarian Carcinoma: A Multicenter Study.

RATIONALE AND OBJECTIVES: This study aims to explore the feasibility of MRI-based habitat radiomics for predicting response of platinum-based chemotherapy in patients with high-grade serous ovarian carcinoma (HGSOC), and compared to conventional radiomics and deep learning models. MATERIALS AND METHODS: A retrospective study was conducted on HGSOC patients from three hospitals. K-means algorithm was used to perform clustering on T2-weighted images (T2WI), contrast-enhanced T1-weighted images (CE-T1WI), and apparent diffusion coefficient (ADC) maps. After feature extraction and selection, the radiomics model, habitat model, and deep learning model were constructed respectively to identify platinum-resistant and platinum-sensitive patients. A nomogram was developed by integrating the optimal model and clinical independent predictors. The model performance and benefit was assessed using the area under the receiver operating characteristic curve (AUC), net reclassification index (NRI), and integrated discrimination improvement (IDI). RESULTS: A total of 394 eligible patients were incorporated. Three habitats were clustered, a significant difference in habitat 2 (weak enhancement, high ADC values, and moderate T2WI signal) was found between the platinum-resistant and platinum-sensitive groups (P < 0.05). Compared to the radiomics model (0.640) and deep learning model (0.603), the habitat model had a higher AUC (0.710). The nomogram, combining habitat signatures with a clinical independent predictor (neoadjuvant chemotherapy), yielded a highest AUC (0.721) among four models, with positive NRI and IDI. CONCLUSION: MRI-based habitat radiomics had the potential to predict response of platinum-based chemotherapy in patients with HGSOC. The nomogram combining with habitat signature had a best performance and good model gains for identifying platinum-resistant patients.

Diffusion kurtosis imaging combined with dynamic susceptibility contrast-enhanced MRI in differentiating high-grade glioma recurrence from pseudoprogression.

OBJECTIVES: To compare the added value of diffusion kurtosis imaging (DKI) with the combination of dynamic susceptibility contrast-enhanced (DSC) MRI in differentiating glioma recurrence from pseudoprogression. METHODS: Thirty-four patients with high-grade gliomas developing new and/or increasing enhanced lesions within six months after surgery and chemoradiotherapy were retrospectively analyzed. All patients were pathologically confirmed to have recurrent glioma (n = 22) or pseudoprogression (n = 12). The DKI and DSC MRI parameters were calculated based on the enhanced lesions on contrast-enhanced T1WI. ROC analysis was performed on significant variables to determine their diagnostic performance. Multivariate logistic regression was used to determine the best prediction model for discrimination. RESULTS: The relative mean kurtosis (rMK), relative axial kurtosis (rKa), relative cerebral blood volume (rCBV), and relative mean transit time (rMTT) of glioma recurrence were higher than those of pseudoprogression (all, P < 0.05). The AUCs and diagnostic accuracy were 0.879 and 82.35% for rMK, 0.723 and 70.59% for rKa, 0.890 and 82.35% for rCBV, 0.765 and 73.53% for rMTT, respectively. A multivariate logistic regression model showed a significant contribution of rMK (P = 0.006) and rCBV (P = 0.009) as independent imaging classifiers for discrimination. The combined use of rMK and rCBV improved the AUC to 0.924 (P < 0.001) and the diagnostic accuracy to 88.24%. CONCLUSION: DKI may be a valuable non-invasive tool in differentiating glioma recurrence from pseudoprogression, and its use in combination with DSC MRI can improve diagnostic performance in assessing treatment response compared with either technique alone.

Hydatidiform mole in a scar on the uterus: A case report.

BACKGROUND: Cesarean scar molar pregnancy is extremely rare, but the incidence has been rising due to the continuous increase in the rate of cesarean section. The presence of a hydatidiform mole in the scar left on the uterus by the procedure may lead to severe complications. We performed a literature review and found only seven reported cases of cesarean scar molar pregnancy. Accurate diagnosis and appropriate treatment are extremely important for the patients' prognosis. CASE SUMMARY: A 35-year-old woman, gravida 4, para 1, complained of vaginal bleeding lasting more than 1 mo and amenorrhea lasting more than 2 mo. The patient's serum human chorionic gonadotropin was 4287800 IU/L. Ultrasound showed a 11.5 cm × 7.5 cm mass at the anterior lower wall of the uterus. The patient underwent suction evacuation, and partial grape-like tissue mixed with blood clots was removed. Uterine arterial embolization was performed to control intraoperative and postoperative bleeding. Histological examination confirmed the presence of a hydatidiform mole in uterine scar. After surgery, there was still a mass with heterogeneous intensity near the isthmus of the uterus on magnetic resonance imaging. The patient then underwent chemotherapy. During the 6-mo follow-up period, the mass disappeared and the serum human chorionic gonadotropin level gradually decreased to normal level. CONCLUSION: We report a case of cesarean scar molar pregnancy successfully cured by comprehensive treatment. We found that cesarean scar molar pregnancy was subject to intraoperative bleeding, and uterine arterial embolization before surgery may be helpful.

Determination of glomerular filtration rate (GFR) from fractional renal accumulation of iodinated contrast material: a convenient and rapid single-kidney CT-GFR technique.

OBJECTIVES: To develop a convenient and rapid single-kidney CT-GFR technique. METHODS: One hundred and twelve patients referred for multiphasic renal CT and 99mTc-DTPA renal dynamic imaging Gates-GFR measurement were prospectively included and randomly divided into two groups of 56 patients each: the training group and the validation group. On the basis of the nephrographic phase images, the fractional renal accumulation (FRA) was calculated and correlated with the Gates-GFR in the training group. From this correlation a formula was derived for single-kidney CT-GFR calculation, which was validated by a paired t test and linear regression analysis with the single-kidney Gates-GFR in the validation group. RESULTS: In the training group, the FRA (x-axis) correlated well (r = 0.95, p < 0.001) with single-kidney Gates-GFR (y-axis), producing a regression equation of y = 1665x + 1.5 for single-kidney CT-GFR calculation. In the validation group, the difference between the methods of single-kidney GFR measurements was 0.38 ± 5.57 mL/min (p = 0.471); the regression line is identical to the diagonal (intercept = 0 and slope = 1) (p = 0.727 and p = 0.473, respectively), with a standard deviation of residuals of 5.56 mL/min. CONCLUSION: A convenient and rapid single-kidney CT-GFR technique was presented and validated in this investigation. KEY POINTS: • The new CT-GFR method takes about 2.5 min of patient time. • The CT-GFR method demonstrated identical results to the Gates-GFR method. • The CT-GFR method is based on the fractional renal accumulation of iodinated CM. • The CT-GFR method is achieved without additional radiation dose to the patient.