Association of Localizing 18F-FDG-PET Hypometabolism and Outcome Following Epilepsy Surgery: Systematic Review and Meta-Analysis.

BACKGROUND AND OBJECTIVES: Although commonly used in the evaluation of patients for epilepsy surgery, the association between the detection of localizing 18fluorine fluorodeoxyglucose PET (18F-FDG-PET) hypometabolism and epilepsy surgery outcome is uncertain. We conducted a systematic review and meta-analysis to determine whether localizing 18F-FDG-PET hypometabolism is associated with favorable outcome after epilepsy surgery. METHODS: A systematic literature search was undertaken. Eligible publications included evaluation with 18F-FDG-PET before epilepsy surgery, with ≥10 participants, and those that reported surgical outcome at ≥12 months. Random-effects meta-analysis was used to calculate the odds of achieving a favorable outcome, defined as Engel class I, International League Against Epilepsy class 1-2, or seizure-free, with localizing 18F-FDG-PET hypometabolism, defined as concordant with the epilepsy surgery resection zone. Meta-regression was used to characterize sources of heterogeneity. RESULTS: The database search identified 8,916 studies, of which 98 were included (total patients n = 4,104). Localizing 18F-FDG-PET hypometabolism was associated with favorable outcome after epilepsy surgery for all patients with odds ratio (OR) 2.68 (95% CI 2.08-3.45). Subgroup analysis yielded similar findings for those with (OR 2.64, 95% CI 1.54-4.52) and without epileptogenic lesion detected on MRI (OR 2.49, 95% CI 1.80-3.44). Concordance with EEG (OR 2.34, 95% CI 1.43-3.83), MRI (OR 1.69, 95% CI 1.19-2.40), and triple concordance with both (OR 2.20, 95% CI 1.32-3.64) was associated with higher odds of favorable outcome. By contrast, diffuse 18F-FDG-PET hypometabolism was associated with worse outcomes compared with focal hypometabolism (OR 0.34, 95% CI 0.22-0.54). DISCUSSION: Localizing 18F-FDG-PET hypometabolism is associated with favorable outcome after epilepsy surgery, irrespective of the presence of an epileptogenic lesion on MRI. The extent of 18F-FDG-PET hypometabolism provides additional information, with diffuse hypometabolism associated with worse surgical outcome than focal 18F-FDG-PET hypometabolism. These findings support the incorporation of 18F-FDG-PET into routine noninvasive investigations for patients being evaluated for epilepsy surgery to improve epileptogenic zone localization and to aid patient selection for surgery.

Identifying epilepsy surgery referral candidates with natural language processing in an Australian context.

OBJECTIVE: Epilepsy surgery is known to be underutilized. Machine learning-natural language processing (ML-NLP) may be able to assist with identifying patients suitable for referral for epilepsy surgery evaluation. METHODS: Data were collected from two tertiary hospitals for patients seen in neurology outpatients for whom the diagnosis of "epilepsy" was mentioned. Individual case note review was undertaken to characterize the nature of the diagnoses discussed in these notes, and whether those with epilepsy fulfilled prespecified criteria for epilepsy surgery workup (namely focal drug refractory epilepsy without contraindications). ML-NLP algorithms were then developed using fivefold cross-validation on the first free-text clinic note for each patient to identify these criteria. RESULTS: There were 457 notes included in the study, of which 250 patients had epilepsy. There were 37 (14.8%) individuals who fulfilled the prespecified criteria for epilepsy surgery referral without described contraindications, 32 (12.8%) of whom were not referred for epilepsy surgical evaluation in the given clinic visit. In the prediction of suitability for epilepsy surgery workup using the prespecified criteria, the tested models performed similarly. For example, the random forest model returned an area under the receiver operator characteristic curve of 0.97 (95% confidence interval 0.93-1.0) for this task, sensitivity of 1.0, and specificity of 0.93. SIGNIFICANCE: This study has shown that there are patients in tertiary hospitals in South Australia who fulfill prespecified criteria for epilepsy surgery evaluation who may not have been referred for such evaluation. ML-NLP may assist with the identification of patients suitable for such referral. PLAIN LANGUAGE SUMMARY: Epilepsy surgery is a beneficial treatment for selected individuals with drug-resistant epilepsy. However, it is vastly underutilized. One reason for this underutilization is a lack of prompt referral of possible epilepsy surgery candidates to comprehensive epilepsy centers. Natural language processing, coupled with machine learning, may be able to identify possible epilepsy surgery candidates through the analysis of unstructured clinic notes. This study, conducted in two tertiary hospitals in South Australia, demonstrated that there are individuals who fulfill criteria for epilepsy surgery evaluation referral but have not yet been referred. Machine learning-natural language processing demonstrates promising results in assisting with the identification of such suitable candidates in Australia.

Identifying epilepsy surgery candidates with natural language processing: A systematic review.

INTRODUCTION: Epilepsy surgery is an underutilised, efficacious management strategy for selected individuals with drug-resistant epilepsy. Natural language processing (NLP) may aid in the identification of patients who are suitable to undergo evaluation for epilepsy surgery. The feasibility of this approach is yet to be determined. METHOD: In accordance with the PRISMA guidelines, a systematic review of the databases PubMed, EMBASE and Cochrane library was performed. This systematic review was prospectively registered on PROSPERO. RESULTS: 6 studies fulfilled inclusion criteria. The majority of included studies reported on datasets from only a single centre, with one study utilising data from two centres and one study six centres. The most commonly employed algorithms were support vector machines (5/6), with only one study utilising NLP strategies such as random forest models and gradient boosted machines. However, the results are promising, with all studies demonstrating moderate to high levels of performance in the identification of patients who may be suitable to undergo epilepsy surgery evaluation. Furthermore, multiple studies demonstrated that NLP could identify such patients 1-2 years prior to the treating clinicians instigating referral. However, no studies were identified that have evaluated the influence of implementing such algorithms on healthcare systems or patient outcomes. CONCLUSIONS: NLP is a promising approach to aid in the identification of patients that may be suitable to undergo epilepsy surgery evaluation. Further studies are required examining diverse datasets with additional analytical methodologies. Studies evaluating the impact of implementation of such algorithms would be beneficial.

18F-FDG-PET hypometabolism as a predictor of favourable outcome in epilepsy surgery: protocol for a systematic review and meta-analysis.

INTRODUCTION: A substantial proportion of patients who undergo surgery for drug resistant focal epilepsy do not become seizure free. While some factors, such as the detection of hippocampal sclerosis or a resectable lesion on MRI and electroencephalogram-MRI concordance, can predict favourable outcomes in epilepsy surgery, the prognostic value of the detection of focal hypometabolism with 18F-fluorodeoxyglucose positive emission tomography (18F-FDG-PET) hypometabolism is uncertain. We propose a protocol for a systematic review and meta-analysis to examine whether localisation with 18F-FDG-PET hypometabolism predicts favourable outcomes in epilepsy surgery. METHODS AND ANALYSIS: A systematic literature search of Medline, Embase and Web of Science will be undertaken. Publications which include evaluation with 18F-FDG-PET prior to surgery for drug resistant focal epilepsy, and which report ≥12 months of postoperative surgical outcome data will be included. Non-human, non-English language publications, publications with fewer than 10 participants and unpublished data will be excluded. Screening and full-text review of publications for inclusion will be undertaken by two independent investigators, with discrepancies resolved by consensus or a third investigator. Data will be extracted and pooled using random effects meta-analysis, with heterogeneity quantified using the I2 analysis. ETHICS AND DISSEMINATION: Ethics approval is not required. Once complete, the systematic review will be published in a peer-reviewed journal. PROSPERO REGISTRATION NUMBER: CRD42022324823.