RESUMO
BACKGROUND: In ischaemic stroke patients undergoing reperfusion therapy, the amount of salvageable tissue, that is, extent of the ischaemic penumbra, predicts the clinical outcomes. CT perfusion (CTP) enables quantification of penumbral tissues to guide decision making, and current programmes have automated its analysis. More advanced machine learning techniques utilising the CTP maps may improve prediction beyond the ischaemic volume measures. METHOD: We determined whether applying convolutional neural networks (CNN), a key machine learning technique in modelling image-label relationships, to post-processed CTP maps improved prediction of outcome, assessed by 3 months modified Rankin scale (mRS). Patients who underwent thrombolysis but not thrombectomy were included. CTP maps of a retrospective cohort of 230 patients with middle cerebral artery stroke were used to develop the model, which was validated in an independent cohort of 129 patients. RESULTS: We constructed a CNN model that predicted a favourable post-thrombolysis outcome (mRS 0-2 at 3 months) with an area under receiver-operator characteristics curve (AUC) of 0.792 (95% CI, 0.707-0.877). This model outperformed a currently clinically used MISTAR software using previously validated thresholds (AUC = 0.583, 95% CI, 0.480-0.686) and a model modified using thresholds from the derivation cohort (AUC = 0.670, 95% CI, 0.571-0.769). By combining CNN-derived features and baseline demographic features, the prediction AUC was improved to 0.865 (95% CI, 0.794-0.936). CONCLUSION: CNN improved prediction of post-thrombolysis outcome, and may be useful in selecting which patients benefit from thrombolysis.
Assuntos
Isquemia Encefálica , Acidente Vascular Cerebral , Humanos , Acidente Vascular Cerebral/diagnóstico por imagem , Isquemia Encefálica/diagnóstico por imagem , Estudos Retrospectivos , Tomografia Computadorizada por Raios X/métodos , Terapia Trombolítica , Imagem de Perfusão/métodosRESUMO
Urethane-based test objects are routinely used for ultrasound quality assurance because of their durability and robustness. The acoustic properties of these phantoms including speed of sound and attenuation, however, have a strong dependence on temperature. Reliable measurement of low-contrast penetration, which is widely used for ultrasound system quality assurance testing, with these phantoms is therefore problematic. To alleviate this, a correction method was proposed using speed of sound estimated by measuring filament target separation. The method was developed using a range of 17 transducer geometry and frequency combinations across 5 ultrasound systems and validated using a further 5 systems. This was found to reduce the uncertainty of low-contrast penetration measurement from an average 17.6 mm to 4.9 mm over the temperature range 8°C to 32°C. This represents a greater than threefold improvement in precision of low-contrast penetration measurement.