A transfer learning nomogram for predicting prostate cancer and benign conditions on MRI.

BACKGROUND: Deep learning has been used to detect or characterize prostate cancer (PCa) on medical images. The present study was designed to develop an integrated transfer learning nomogram (TLN) for the prediction of PCa and benign conditions (BCs) on magnetic resonance imaging (MRI). METHODS: In this retrospective study, a total of 709 patients with pathologically confirmed PCa and BCs from two institutions were included and divided into training (n = 309), internal validation (n = 200), and external validation (n = 200) cohorts. A transfer learning signature (TLS) that was pretrained with the whole slide images of PCa and fine-tuned on prebiopsy MRI images was constructed. A TLN that integrated the TLS, the Prostate Imaging-Reporting and Data System (PI-RADS) score, and the clinical factor was developed by multivariate logistic regression. The performance of the TLS, clinical model (CM), and TLN were evaluated in the validation cohorts using the receiver operating characteristic (ROC) curve, the Delong test, the integrated discrimination improvement (IDI), and decision curve analysis. RESULTS: TLS, PI-RADS score, and age were selected for TLN construction. The TLN yielded areas under the curve of 0.9757 (95% CI, 0.9613-0.9902), 0.9255 (95% CI, 0.8873-0.9638), and 0.8766 (95% CI, 0.8267-0.9264) in the training, internal validation, and external validation cohorts, respectively, for the discrimination of PCa and BCs. The TLN outperformed the TLS and the CM in both the internal and external validation cohorts. The decision curve showed that the TLN added more net benefit than the CM. CONCLUSIONS: The proposed TLN has the potential to be used as a noninvasive tool for PCa and BCs differentiation.

Case report: Leptomeningeal metastasis of advanced nasopharyngeal carcinoma treated with chemoimmunotherapy.

Leptomeningeal metastasis (LM) of nasopharyngeal carcinoma (NPC) is rare and associated with a poor prognosis. Immune checkpoint inhibitors (ICIs) have been the standard first-line treatment for metastatic NPC, but their effect on meningeal metastasis of NPC needs further investigation. A 38-year-old man complained of bilateral neck masses and sought medical care. He was diagnosed with nasopharyngeal undifferentiated non-keratinizing carcinoma with bilateral cervical lymph node metastasis and multiple bone metastasis, stage cT4N2M1 IVb. Then, the patient received first-line anti-PD-1 antibody tislelizumab combined with gemcitabine and cisplatin and achieved partial response. After seven cycles of first-line chemoimmunotherapy, the patient subsequently developed neurological symptoms, including unsteady walking, slurred speech, coughing on drinking, and unconsciousness. MRI showed leptomeningeal linear enhancement, and cerebrospinal fluid (CSF) analysis indicated Epstein-Barr virus (EBV) infection and squamous cell carcinoma cytology, suggesting the diagnosis of leptomeningeal metastasis. After the definite diagnosis of LM, the patient's condition deteriorated rapidly, leading to his death from brain herniation. We reported the first case of advanced NPC with pathologically confirmed leptomeningeal metastasis after receiving first-line chemoimmunotherapy. Considering the poor prognosis of LM, it is suggested to perform MRI and CSF examination when patients have neurological symptoms. Although immunotherapy significantly improved survival outcomes of advanced NPC patients, it seemed not effective in the setting of LM. The effect of other treatment options, such as radiation therapy and intrathecal therapy, requires further verification.

Combination of Quantitative Susceptibility Mapping and Diffusion Kurtosis Imaging Provides Potential Biomarkers for Early-Stage Parkinson's Disease.

Purpose: This study aimed to detect changes in iron deposition and neural microstructure in the substantia nigra (SN), red nucleus (RN), and basal ganglia of Parkinson's disease (PD) patients at different stages using quantitative susceptibility mapping and diffusion kurtosis imaging to identify potential indicators of early-stage PD. Methods: We enrolled 20 early-stage and 15 late-stage PD patients, as well as 20 age- and sex-matched controls. All participants underwent quantitative susceptibility mapping and diffusion kurtosis imaging to determine magnetic susceptibility (MS), fractional anisotropy (FA), mean diffusivity (MD), and mean kurtosis (MK) in several brain regions. Results: Compared with the control group, MS and MK values in the SN were significantly increased in the early- and late-stage PD group, whereas MS values in the red nucleus (RN), globus pallidus (GP), and caudate nucleus (CN), FA value in the CN and GP, and MK value in the CN and putamen (PU) were significantly increased in the late-stage PD group. There were positive correlations between MS and MK values in the CN and MS and FA values in the GP. Furthermore, the combination of MS and MK values in the SN provided high accuracy for distinguishing early-stage PD patients from controls. Conclusions: This study identified MS and MK in the SN as potential indicators of early-stage PD.

Diffusion Kurtosis Imaging for Detection of Early Brain Changes in Parkinson's Disease.

We aimed to evaluate microscale changes in the bilateral red nucleus and substantia nigra of patients with Parkinson's disease (PD) using diffusion kurtosis imaging (DKI). Twenty-six patients with PD [mean age, 62.5 ± 8.7 years; Hoehn-Yahr stage, 0-4.0; Unified Parkinson's Disease Rating Scale (UPDRS) scores, 8-43] and 15 healthy controls (mean age, 59.5 ± 9.4 years) underwent DKI of the substantia nigra and red nucleus. Imaging was performed using a General Electric (GE) Signa 3.0-T MRI system. Patients with PD were divided into two groups consisting of 12 patients with UPDRS scores ≥ 30 and 14 patients with UPDRS scores < 30. All DKI data processing operations were performed with commercial workstations (GE, ADW 4.6) using Functool software to generate color-coded and parametric maps of mean kurtosis (MK), fractional anisotropy (FA), and mean diffusivity (MD). MK values in the bilateral substantia nigra were significantly lower in patients with early- and advanced-stage PD than in controls. Moreover, MK values in the left substantia nigra were significantly lower in patients with advanced-stage PD than in those with early-stage PD. Patients with advanced-stage PD also exhibited significant decreases in MK values in the bilateral red nucleus relative to controls. No significant differences in FA or MD values were observed between the PD and control groups. There were no significant correlations between MK, FA, or MD values and UPDRS scores. Our findings suggest that decreased MK values in the substantia nigra may aid in determining the severity of PD and help provide early diagnoses.

Extracellular pH is a biomarker enabling detection of breast cancer and liver cancer using CEST MRI.

Extracellular pH (pHe) decrease is associated with tumor growth, invasion, metastasis, and chemoresistance, which can be detected by chemical exchange saturation transfer (CEST) magnetic resonance imaging (MRI). Here, we demonstrated that ioversol CEST MRI can be exploited to achieve pHe mapping of the liver cancer microenvironment. In in vitro studies, we firstly explored whether ioversol signal is pH-dependent, and calculated the function equation between the CEST effects of ioversol and pH values, in the range of 6.0 to 7.8, by a ratiometric method. Then we verified the feasibility of this technique and the equation in vivo by applying pHe imaging in an MMTV-Erbb2 transgenic mouse breast cancer model, which is often used in CEST pHe studies. Furthermore, in vivo ioversol CEST MRI, we were able to map relative pHe and differentiate between tumor and normal tissue in a McA-RH7777 rat hepatoma model. This suggests pHe may be a useful biomarker for human liver cancer.