Predicting Response to Neoadjuvant Chemotherapy with PET Imaging Using Convolutional Neural Networks

Clinical utility of texture analysis of 18F-FDG PET/CT in patients with Stage I lung cancer treated with stereotactic body radiotherapy

We evaluated the reproducibility and predictive value of texture parameters and existing parameters of 18F-FDG PET/CT images in Stage I non-small-cell lung cancer (NSCLC) patients treated with stereotactic body radiotherapy (SBRT). Twenty-six patients with Stage I NSCLC (T1-2N0M0) were retrospectively analyzed. All of the patients underwent an 18F-FDG PET/CT scan before treatment and were treated with SBRT. Each tumor was delineated using PET Edge (MIM Software Inc., Cleveland, OH), and texture parameters were calculated using open-source code CGITA. From 18F-FDG PET/CT images, three conventional parameters, including maximum standardized uptake value (SUV), metabolic tumor volume (MTV) and total lesion glycolysis (TLG), and four texture parameters, including entropy and dissimilarity (derived from a co-occurrence matrix) and high-intensity large-area emphasis (HILAE) and zone percentage (derived from a size-zone matrix) were analyzed. Reproducibility was evaluated using two independent delineations conducted by two observers. The ability to predict local control (LC), progression-free survival (PFS) and overall survival (OS) was tested for each parameter. All of the seven parameters except zone percentage showed good reproducibility, with intraclass correlation coefficient values >0.8. In univariate analysis, only HILAE was a significant predictor for LC. Histology, dose fractionation, and maximum SUV were associated with PFS, and histology and dose fractionation were associated with OS. We showed that texture parameters derived from 18F-FDG PET/CT were reproducible and potentially beneficial for predicting LC in Stage I lung cancer patients treated with SBRT.

FDG PET/CT radiomics for predicting the outcome of locally advanced rectal cancer

PURPOSE

The aim of this study was to investigate the prognostic value of baseline ¹⁸F-FDG PET/CT textural analysis in locally-advanced rectal cancer (LARC).

METHODS

Eighty-six patients with LARC underwent ¹⁸F-FDG PET/CT before treatment. Maximum and mean standard uptake values (SUVmax and SUVmean), metabolic tumoral volume (MTV), total lesion glycolysis (TLG), histogram-intensity features, as well as 11 local and regional textural features, were evaluated. The relationships of clinical, pathological and PET-derived metabolic parameters with disease-specific survival (DSS), disease-free survival (DFS) and overall survival (OS) were assessed by Cox regression analysis. Logistic regression was used to predict the pathological response by the Dworak tumor regression grade (TRG) in the 66 patients treated with neoadjuvant chemoradiotherapy (nCRT).

RESULTS

The median follow-up of patients was 41 months. Seventeen patients (19.7%) had recurrent disease and 18 (20.9 %) died, either due to cancer progression (n = 10) or from another cause while in complete remission (n = 8). DSS was 95% at 1 year, 93% at 2 years and 87% at 4 years. Weight loss, surgery and the texture parameter coarseness were significantly associated with DSS in multivariate analyses. DFS was 94 % at 1 year, 86 % at 2 years and 79 % at 4 years. From a multivariate standpoint, tumoral differentiation and the texture parameters homogeneity and coarseness were significantly associated with DFS. OS was 93% at 1 year, 87% at 2 years and 79% after 4 years. cT, surgery, SUVmean, dissimilarity and contrast from the neighborhood intensity-difference matrix (contrast_NGTDM) were significantly and independently associated with OS. Finally, RAS-mutational status (KRAS and NRAS mutations) and TLG were significant predictors of pathological response to nCRT (TRG 3-4).

CONCLUSION

Textural analysis of baseline ¹⁸F-FDG PET/CT provides strong independent predictors of survival in patients with LARC, with better predictive power than intensity- and volume-based parameters. The utility of such features, especially coarseness, should be confirmed by larger clinical studies before considering their potential integration into decisional algorithms aimed at personalized medicine.

Automated Critical Test Findings Identification and Online Notification System Using Artificial Intelligence in Imaging

Purpose To evaluate the performance of an artificial intelligence (AI) tool using a deep learning algorithm for detecting hemorrhage, mass effect, or hydrocephalus (HMH) at non-contrast material-enhanced head computed tomographic (CT) examinations and to determine algorithm performance for detection of suspected acute infarct (SAI). Materials and Methods This HIPAA-compliant retrospective study was completed after institutional review board approval. A training and validation dataset of noncontrast-enhanced head CT examinations that comprised 100 examinations of HMH, 22 of SAI, and 124 of noncritical findings was obtained resulting in 2583 representative images. Examinations were processed by using a convolutional neural network (deep learning) using two different window and level configurations (brain window and stroke window). AI algorithm performance was tested on a separate dataset containing 50 examinations with HMH findings, 15 with SAI findings, and 35 with noncritical findings. Results Final algorithm performance for HMH showed 90% (45 of 50) sensitivity (95% confidence interval [CI]: 78%, 97%) and 85% (68 of 80) specificity (95% CI: 76%, 92%), with area under the receiver operating characteristic curve (AUC) of 0.91 with the brain window. For SAI, the best performance was achieved with the stroke window showing 62% (13 of 21) sensitivity (95% CI: 38%, 82%) and 96% (27 of 28) specificity (95% CI: 82%, 100%), with AUC of 0.81. Conclusion AI using deep learning demonstrates promise for detecting critical findings at noncontrast-enhanced head CT. A dedicated algorithm was required to detect SAI. Detection of SAI showed lower sensitivity in comparison to detection of HMH, but showed reasonable performance. Findings support further investigation of the algorithm in a controlled and prospective clinical setting to determine whether it can independently screen noncontrast-enhanced head CT examinations and notify the interpreting radiologist of critical findings. ^© RSNA, 2017 Online supplemental material is available for this article.

Beyond imaging: The promise of radiomics

The domain of investigation of radiomics consists of large-scale radiological image analysis and association with biological or clinical endpoints. The purpose of the present study is to provide a recent update on the status of this rapidly emerging field by performing a systematic review of the literature on radiomics, with a primary focus on oncologic applications. The systematic literature search, performed in Pubmed using the keywords: "radiomics OR radiomic" provided 97 research papers. Based on the results of this search, we describe the methods used for building a model of prognostic value from quantitative analysis of patient images. Then, we provide an up-to-date overview of the results achieved in this field, and discuss the current challenges and future developments of radiomics for oncology.

Localization and diagnosis framework for pediatric cataracts based on slit-lamp images using deep features of a convolutional neural network

Slit-lamp images play an essential role for diagnosis of pediatric cataracts. We present a computer vision-based framework for the automatic localization and diagnosis of slit-lamp images by identifying the lens region of interest (ROI) and employing a deep learning convolutional neural network (CNN). First, three grading degrees for slit-lamp images are proposed in conjunction with three leading ophthalmologists. The lens ROI is located in an automated manner in the original image using two successive applications of Candy detection and the Hough transform, which are cropped, resized to a fixed size and used to form pediatric cataract datasets. These datasets are fed into the CNN to extract high-level features and implement automatic classification and grading. To demonstrate the performance and effectiveness of the deep features extracted in the CNN, we investigate the features combined with support vector machine (SVM) and softmax classifier and compare these with the traditional representative methods. The qualitative and quantitative experimental results demonstrate that our proposed method offers exceptional mean accuracy, sensitivity and specificity: classification (97.07%, 97.28%, and 96.83%) and a three-degree grading area (89.02%, 86.63%, and 90.75%), density (92.68%, 91.05%, and 93.94%) and location (89.28%, 82.70%, and 93.08%). Finally, we developed and deployed a potential automatic diagnostic software for ophthalmologists and patients in clinical applications to implement the validated model.

Characterization of PET/CT images using texture analysis: the past, the present… any future?

After seminal papers over the period 2009 - 2011, the use of texture analysis of PET/CT images for quantification of intratumour uptake heterogeneity has received increasing attention in the last 4 years. Results are difficult to compare due to the heterogeneity of studies and lack of standardization. There are also numerous challenges to address. In this review we provide critical insights into the recent development of texture analysis for quantifying the heterogeneity in PET/CT images, identify issues and challenges, and offer recommendations for the use of texture analysis in clinical research. Numerous potentially confounding issues have been identified, related to the complex workflow for the calculation of textural features, and the dependency of features on various factors such as acquisition, image reconstruction, preprocessing, functional volume segmentation, and methods of establishing and quantifying correspondences with genomic and clinical metrics of interest. A lack of understanding of what the features may represent in terms of the underlying pathophysiological processes and the variability of technical implementation practices makes comparing results in the literature challenging, if not impossible. Since progress as a field requires pooling results, there is an urgent need for standardization and recommendations/guidelines to enable the field to move forward. We provide a list of correct formulae for usual features and recommendations regarding implementation. Studies on larger cohorts with robust statistical analysis and machine learning approaches are promising directions to evaluate the potential of this approach.

The emerging field of radiomics in esophageal cancer: current evidence and future potential

'Radiomics' is the name given to the emerging field of extracting additional information from standard medical images using advanced feature analysis. This innovative form of quantitative image analysis appears to have future potential for clinical practice in patients with esophageal cancer by providing an additional layer of information to the standard imaging assessment. There is a growing body of evidence suggesting that radiomics may provide incremental value for staging, predicting treatment response, and predicting survival in esophageal cancer, for which the current work-up has substantial limitations. This review outlines the available evidence and future potential for the application of radiomics in the management of patients with esophageal cancer. In addition, an overview of the current evidence on the importance of reproducibility of image features and the substantial influence of varying smoothing scales, quantization levels, and segmentation methods is provided.

Deep Learning in Drug Discovery

Artificial neural networks had their first heyday in molecular informatics and drug discovery approximately two decades ago. Currently, we are witnessing renewed interest in adapting advanced neural network architectures for pharmaceutical research by borrowing from the field of "deep learning". Compared with some of the other life sciences, their application in drug discovery is still limited. Here, we provide an overview of this emerging field of molecular informatics, present the basic concepts of prominent deep learning methods and offer motivation to explore these techniques for their usefulness in computer-assisted drug discovery and design. We specifically emphasize deep neural networks, restricted Boltzmann machine networks and convolutional networks.