Usefulness of Histogram-Profile Analysis in Ring-Enhancing Intracranial Lesions

Diffusion-weighted imaging-based radiomics model using automatic machine learning to differentiate cerebral cystic metastases from brain abscesses

OBJECTIVES

To develop a radiomics model based on diffusion-weighted imaging (DWI) utilizing automated machine learning method to differentiate cerebral cystic metastases from brain abscesses.

MATERIALS AND METHODS

A total of 186 patients with cerebral cystic metastases (n = 98) and brain abscesses (n = 88) from two clinical institutions were retrospectively included. The datasets (129 from institution A) were randomly portioned into separate 75% training and 25% internal testing sets. Radiomics features were extracted from DWI images using two subregions of the lesion (cystic core and solid wall). A thorough image preprocessing method was applied to DWI images to ensure the robustness of radiomics features before feature extraction. Then the Tree-based Pipeline Optimization Tool (TPOT) was utilized to search for the best optimized machine learning pipeline, using a fivefold cross-validation in the training set. The external test set (57 from institution B) was used to evaluate the model's performance.

RESULTS

Seven distinct TPOT models were optimized to distinguish between cerebral cystic metastases and abscesses either based on different features combination or using wavelet transform. The optimal model demonstrated an AUC of 1.00, an accuracy of 0.97, sensitivity of 1.00, and specificity of 0.93 in the internal test set, based on the combination of cystic core and solid wall radiomics signature using wavelet transform. In the external test set, this model reached 1.00 AUC, 0.96 accuracy, 1.00 sensitivity, and 0.93 specificity.

CONCLUSION

The DWI-based radiomics model established by TPOT exhibits a promising predictive capacity in distinguishing cerebral cystic metastases from abscesses.

Independent risk factors associated with higher mortality rates and recurrence of brain abscesses from head and neck sources

OBJECTIVE

The aim of this study was to identify the features and independent risk factors associated with recurrence and mortality in patients with brain abscesses of head and neck origin.

STUDY DESIGN

We designed a retrospective study of patients diagnosed with a brain abscess at the Massachusetts General Hospital between 1980 and 2017. Inclusion criteria were complete medical records, including medical and surgical history; and radiographic and microbiologic data. Multinomial logistic regression and Gray's test were used to evaluate the independent variables associated with recurrence and mortality.

RESULTS

Eighty-eight cases met the inclusion criteria. Of these, 48 patients (54.5%) were men (mean age 50.5 ± 18.8 years). Significant association between etiology and cultured organisms was found only in cases of neurosurgical intervention with staphylococcal or streptococcal isolates (P < .05). Seizure activity was the only significant predictor of recurrence. Predictors of mortality included advanced age (P = .005); staphylococcal infection (P = .029); low monocyte count (P = .004); hyponatremia (P = .002); elevated blood urea nitrogen (P = .000); elevated creatinine (P = .002); hyperglycemia (P = .023); and status at discharge (P = .000).

CONCLUSIONS

Independent risk factors, such as low monocyte count, hyponatremia, renal dysfunction, and hyperglycemia, were found to be associated with higher mortality rates in patients with brain abscesses of head and neck origin. These abnormalities should be promptly recognized and aggressively treated.