Development and validation of a machine learning model to predict prognosis in HIV-negative cryptococcal meningitis patients: a multicenter study.

PURPOSE: To predict prognosis in HIV-negative cryptococcal meningitis (CM) patients by developing and validating a machine learning (ML) model. METHODS: This study involved 523 HIV-negative CM patients diagnosed between January 1, 1998, and August 31, 2022, by neurologists from 3 tertiary Chinese centers. Prognosis was evaluated at 10 weeks after the initiation of antifungal therapy. RESULTS: The final prediction model for HIV-negative CM patients comprised 8 variables: Cerebrospinal fluid (CSF) cryptococcal count, CSF white blood cell (WBC), altered mental status, hearing impairment, CSF chloride levels, CSF opening pressure (OP), aspartate aminotransferase levels at admission, and decreased rate of CSF cryptococcal count within 2 weeks after admission. The areas under the curve (AUCs) in the internal, temporal, and external validation sets were 0.87 (95% CI 0.794-0.944), 0.92 (95% CI 0.795-1.000), and 0.86 (95% CI 0.744-0.975), respectively. An artificial intelligence (AI) model was trained to detect and count cryptococci, and the mean average precision (mAP) was 0.993. CONCLUSION: A ML model for predicting prognosis in HIV-negative CM patients was built and validated, and the model might provide a reference for personalized treatment of HIV-negative CM patients. The change in the CSF cryptococcal count in the early phase of HIV-negative CM treatment can reflect the prognosis of the disease. In addition, utilizing AI to detect and count CSF cryptococci in HIV-negative CM patients can eliminate the interference of human factors in detecting cryptococci in CSF samples and reduce the workload of the examiner.

Diffusion-weighted magnetic resonance imaging of mucin pools in locally advanced rectal mucinous adenocarcinoma predicts tumor response to neoadjuvant therapy.

PURPOSE: To determine the prognostic value of diffusion-weighted magnetic resonance imaging (DW-MRI) of mucin pools (MPs) in predicting the response of patients with locally advanced rectal mucinous adenocarcinoma (RMAC) to neoadjuvant therapy (NAT). METHOD: A total of 59 patients with histologically proven RMAC received NAT before applying total mesorectal excision. MP and solid tumor (ST) components were identified using T2 weighted image (T2WI) and DW-MRI, and apparent diffusion coefficient (ADC) values were calculated prior, during and after NAT. The receiver operating characteristic (ROC) curve was used to evaluate the ability of ADC values in predicting NAT efficacy as determined by post-pathological tumor regression grade (TRG). In addition, radiologists evaluated the TNM staging of tumors, the mesorectal fascia invasion, the maximal tumor length, and the distance from the inferior part of the tumor to the anal verge. Multivariate analysis and logistic regression were used to determine the correlation of ADC values and baseline MRI parameters with NAT efficacy. RESULTS: Among the 59 patients, 44 (74.6 %) were men. The mean age of patients was 49.5 ± 11.2 years. The mean ΔADC value during NAT obtained on mucus pool was higher in the responsiveness group than that of the nonresponsiveness group (0.506 ± 0.342 vs. 0.053 ± 0.240 × 10-3 mm2/s, P < .001), with an area under the curve of receiver operating characteristic of 0.881 (95 %CI, 0.770-0.951). CONCLUSIONS: MRI can be reliably used to measure MP-ADC, which as we showed in this study, represents a biomarker to predict tumor responsiveness of NAT in RMAC patients.