Autonomous screening for laser photocoagulation in fundus images using deep learning.

BACKGROUND: Diabetic retinopathy (DR) is a leading cause of blindness in adults worldwide. Artificial intelligence (AI) with autonomous deep learning algorithms has been increasingly used in retinal image analysis, particularly for the screening of referrable DR. An established treatment for proliferative DR is panretinal or focal laser photocoagulation. Training autonomous models to discern laser patterns can be important in disease management and follow-up. METHODS: A deep learning model was trained for laser treatment detection using the EyePACs dataset. Data was randomly assigned, by participant, into development (n=18 945) and validation (n=2105) sets. Analysis was conducted at the single image, eye, and patient levels. The model was then used to filter input for three independent AI models for retinal indications; changes in model efficacy were measured using area under the receiver operating characteristic curve (AUC) and mean absolute error (MAE). RESULTS: On the task of laser photocoagulation detection: AUCs of 0.981, 0.95, and 0.979 were achieved at the patient, image, and eye levels, respectively. When analysing independent models, efficacy was shown to improve across the board after filtering. Diabetic macular oedema detection on images with artefacts was AUC 0.932 vs AUC 0.955 on those without. Participant sex detection on images with artefacts was AUC 0.872 vs AUC 0.922 on those without. Participant age detection on images with artefacts was MAE 5.33 vs MAE 3.81 on those without. CONCLUSION: The proposed model for laser treatment detection achieved high performance on all analysis metrics and has been demonstrated to positively affect the efficacy of different AI models, suggesting that laser detection can generally improve AI-powered applications for fundus images.

Model-free dynamic contrast-enhanced MRI analysis: differentiation between active tumor and necrotic tissue in patients with glioblastoma.

OBJECTIVE: Treatment response assessment in patients with high-grade gliomas (HGG) is heavily dependent on changes in lesion size on MRI. However, in conventional MRI, treatment-related changes can appear as enhancing tissue, with similar presentation to that of active tumor tissue. We propose a model-free data-driven method for differentiation between these tissues, based on dynamic contrast-enhanced (DCE) MRI. MATERIALS AND METHODS: The study included a total of 66 scans of patients with glioblastoma. Of these, 48 were acquired from 1 MRI vendor and 18 scans were acquired from a different MRI vendor and used as test data. Of the 48, 24 scans had biopsy results. Analysis included semi-automatic arterial input function (AIF) extraction, direct DCE pharmacokinetic-like feature extraction, and unsupervised clustering of the two tissue types. Validation was performed via (a) comparison to biopsy result (b) correlation to literature-based DCE curves for each tissue type, and (c) comparison to clinical outcome. RESULTS: Consistency between the model prediction and biopsy results was found in 20/24 cases. An average correlation of 82% for active tumor and 90% for treatment-related changes was found between the predicted component and population-based templates. An agreement between the predicted results and radiologist's assessment, based on RANO criteria, was found in 11/12 cases. CONCLUSION: The proposed method could serve as a non-invasive method for differentiation between lesion tissue and treatment-related changes.

Differentiation between glioblastoma, brain metastasis and subtypes using radiomics analysis.

BACKGROUND: Differentiation between glioblastoma and brain metastasis is highly important due to differing medical treatment strategies. While MRI is the modality of choice for the assessment of patients with brain tumors, differentiation between glioblastoma and solitary brain metastasis may be challenging due to their similar appearance on MRI. PURPOSE: To differentiate between glioblastoma and brain metastasis subtypes using radiomics analysis based on conventional post-contrast T1 -weighted (T1 W) MRI. STUDY TYPE: Retrospective. SUBJECTS: Data were acquired from 439 patients: 212 patients with glioblastoma and 227 patients with brain metastasis (breast, lung, and others). FIELD STRENGTH/SEQUENCE: Post-contrast 3D T1 W gradient echo images, acquired with 1.5 and 3.0 T MR systems. ASSESSMENT: Analysis included image preprocessing, segmentation of tumor area, and features extraction including: patients' clinical information, tumor location, first- and second-order statistical, morphological, wavelet features, and bag-of-features. Following dimension reduction, classification was performed using various machine-learning algorithms including support-vector machine (SVM), k-nearest neighbor, decision trees, and ensemble classifiers. STATISTICAL TESTS: For classification, the data were divided into training (80%) and testing datasets (20%). Following optimization of the classifiers, mean sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were calculated. RESULTS: For the testing dataset, the best results for differentiation of glioblastoma from brain metastasis were obtained using the SVM classifier with mean accuracy = 0.85, sensitivity = 0.86, specificity = 0.85, and AUC = 0.96. The best classification results between glioblastoma and brain metastasis subtypes were obtained using SVM classifier with mean accuracy = 0.85, 0.89, 0.75, 0.90; sensitivity = 1.00, 0.60, 0.57, 0.11; specificity = 0.76, 0.92, 0.87, 0.99; and AUC = 0.98, 0.81, 0.83, 0.57 for the glioblastoma, breast, lung, and other brain metastases, respectively. DATA CONCLUSION: Differentiation between glioblastoma and brain metastasis showed a high success rate based on postcontrast T1 W MRI. Classification between glioblastoma and brain metastasis subtypes may require additional MR sequences with other tissue contrasts. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;50:519-528.