Robust deep learning-based PET prognostic imaging biomarker for DLBCL patients: a multicenter study.

OBJECTIVE: To develop and independently externally validate robust prognostic imaging biomarkers distilled from PET images using deep learning techniques for precise survival prediction in patients with diffuse large B cell lymphoma (DLBCL). METHODS: A total of 684 DLBCL patients from three independent medical centers were included in this retrospective study. Deep learning scores (DLS) were generated from PET images using deep convolutional neural network architecture known as VGG19 and DenseNet121. These DLSs were utilized to predict progression-free survival (PFS) and overall survival (OS). Furthermore, multiparametric models were designed based on results from the Cox proportional hazards model and assessed through calibration curves, concordance index (C-index), and decision curve analysis (DCA) in the training and validation cohorts. RESULTS: The DLSPFS and DLSOS exhibited significant associations with PFS and OS, respectively (P<0.05) in the training and validation cohorts. The multiparametric models that incorporated DLSs demonstrated superior efficacy in predicting PFS (C-index: 0.866) and OS (C-index: 0.835) compared to competing models in training cohorts. In external validation cohorts, the C-indices for PFS and OS were 0.760 and. 0.770 and 0.748 and 0.766, respectively, indicating the reliable validity of the multiparametric models. The calibration curves displayed good consistency, and the decision curve analysis (DCA) confirmed that the multiparametric models offered more net clinical benefits. CONCLUSIONS: The DLSs were identified as robust prognostic imaging biomarkers for survival in DLBCL patients. Moreover, the multiparametric models developed in this study exhibited promising potential in accurately stratifying patients based on their survival risk.

Prognosis Prediction of Diffuse Large B-Cell Lymphoma in 18F-FDG PET Images Based on Multi-Deep-Learning Models.

Diffuse large B-cell lymphoma (DLBCL), a cancer of B cells, has been one of the most challenging and complicated diseases because of its considerable variation in clinical behavior, response to therapy, and prognosis. Radiomic features from medical images, such as PET images, have become one of the most valuable features for disease classification or prognosis prediction using learning-based methods. In this paper, a new flexible ensemble deep learning model is proposed for the prognosis prediction of the DLBCL in 18F-FDG PET images. This study proposes the multi-R-signature construction through selected pre-trained deep learning models for predicting progression-free survival (PFS) and overall survival (OS). The proposed method is trained and validated on two datasets from different imaging centers. Through analyzing and comparing the results, the prediction models, including Age, Ann abor stage, Bulky disease, SUVmax, TMTV, and multi-R-signature, achieve the almost best PFS prediction performance (C-index: 0.770, 95% CI: 0.705-0.834, with feature adding fusion method and C-index: 0.764, 95% CI: 0.695-0.832, with feature concatenate fusion method) and OS prediction (C-index: 0.770 (0.692-0.848) and 0.771 (0.694-0.849)) on the validation dataset. The developed multiparametric model could achieve accurate survival risk stratification of DLBCL patients. The outcomes of this study will be helpful for the early identification of high-risk DLBCL patients with refractory relapses and for guiding individualized treatment strategies.

Learning-based initialization for correntropy-based level sets to segment atherosclerotic plaque in ultrasound images.

Carotid artery atherosclerosis is a significant cause of stroke. Ultrasound imaging has been widely used in the diagnosis of atherosclerosis. Therefore, segmenting the atherosclerotic carotid plaque in an ultrasound image is an important task. Accurate plaque segmentation is helpful for the measurement of carotid plaque burden. This study proposes an automatic method for atherosclerotic plaque segmentation by using correntropy-based level sets (CLS) with learning-based initialization. We introduce the CLS model, containing the point-based local bias-field corrected image fitting method and correntropy-based distance measurement, to overcome the limitations of the ultrasound images. A supervised learning algorithm is employed to solve the automatic initialization problem of the variational methods. The proposed atherosclerotic plaque segmentation method is validated on 29 carotid ultrasound images, obtaining a Dice ratio of 90.6 ± 1.9% and an overlap index of 83.6 ± 3.2%. Moreover, by comparing the standard deviation of each evaluation index, it can be found that the proposed method is more robust for segmenting the atherosclerotic plaque. Our work shows that our proposed method can be more helpful than other variational models for measuring the carotid plaque burden.

Segmentation of the Common Carotid Intima-Media Complex in Ultrasound Images Using 2-D Continuous Max-Flow and Stacked Sparse Auto-encoder.

The intima-media thickness (IMT) of a common carotid artery in an ultrasound image is considered an important indicator of the onset of atherosclerosis. However, it is challenging to segment the intima-media complex (IMC) directly in ultrasound images. This study proposes a fully automatic method to segment the IMC on longitudinal B-mode ultrasound images. Our method consists of two stages: (i) extraction of the region of interest with a continuous max-flow algorithm and region-of-interest reconstruction using a stacked sparse auto-encoder model, and (ii) IMC segmentation using a trained random forest classifier. The proposed method has been tested on three databases from three different imaging centres, comprising a total of 228 ultrasound images of the common carotid artery. On the three databases, our method yields mean absolute errors of 0.028 ± 0.016 mm, 0.579 ± 0.288 pixel and 0.582 ± 0.341 pixel; polyline distance (PD) measures of 0.026 ± 0.017 mm, 0.657 ± 0.275 pixel and 0.731 ± 0:282 pixel; Hausdorff distance measures of 0.249 ± 0.101 mm, 4.760 ± 1.085 pixels and 5.825 ± 2.059 pixels; and correlation coefficients of 95.19%, 93.79%, and 98.96%, respectively. These results indicate that the proposed method performs well in segmentation of the IMC and measurement of the IMT.

An integrated method for atherosclerotic carotid plaque segmentation in ultrasound image.

BACKGROUND AND OBJECTIVE: Carotid artery atherosclerosis is an important cause of stroke. Ultrasound imaging has been widely used in the diagnosis of atherosclerosis. Therefore, segmenting atherosclerotic carotid plaque in ultrasound image is an important task. Accurate plaque segmentation is helpful for the measurement of carotid plaque burden. In this paper, we propose and evaluate a novel learning-based integrated framework for plaque segmentation. METHODS: In our study, four different classification algorithms, along with the auto-context iterative algorithm, were employed to effectively integrate features from ultrasound images and later also the iteratively estimated and refined probability maps together for pixel-wise classification. The four classification algorithms were support vector machine with linear kernel, support vector machine with radial basis function kernel, AdaBoost and random forest. The plaque segmentation was implemented in the generated probability map. The performance of the four different learning-based plaque segmentation methods was tested on 29 B-mode ultrasound images. The evaluation indices for our proposed methods were consisted of sensitivity, specificity, Dice similarity coefficient, overlap index, error of area, absolute error of area, point-to-point distance, and Hausdorff point-to-point distance, along with the area under the ROC curve. RESULTS: The segmentation method integrated the random forest and an auto-context model obtained the best results (sensitivity 80.4 ±â€¯8.4%, specificity 96.5 ±â€¯2.0%, Dice similarity coefficient 81.0 ±â€¯4.1%, overlap index 68.3 ±â€¯5.8%, error of area -1.02 ±â€¯18.3%, absolute error of area 14.7 ±â€¯10.9%, point-to-point distance 0.34 ±â€¯0.10 mm, Hausdorff point-to-point distance 1.75 ±â€¯1.02 mm, and area under the ROC curve 0.897), which were almost the best, compared with that from the existed methods. CONCLUSIONS: Our proposed learning-based integrated framework investigated in this study could be useful for atherosclerotic carotid plaque segmentation, which will be helpful for the measurement of carotid plaque burden.

MRI-based prostate cancer detection with high-level representation and hierarchical classification.

PURPOSE: Extracting the high-level feature representation by using deep neural networks for detection of prostate cancer, and then based on high-level feature representation constructing hierarchical classification to refine the detection results. METHODS: High-level feature representation is first learned by a deep learning network, where multiparametric MR images are used as the input data. Then, based on the learned high-level features, a hierarchical classification method is developed, where multiple random forest classifiers are iteratively constructed to refine the detection results of prostate cancer. RESULTS: The experiments were carried on 21 real patient subjects, and the proposed method achieves an averaged section-based evaluation (SBE) of 89.90%, an averaged sensitivity of 91.51%, and an averaged specificity of 88.47%. CONCLUSIONS: The high-level features learned from our proposed method can achieve better performance than the conventional handcrafted features (e.g., LBP and Haar-like features) in detecting prostate cancer regions, also the context features obtained from the proposed hierarchical classification approach are effective in refining cancer detection result.

In vivo MRI based prostate cancer localization with random forests and auto-context model.

Prostate cancer is one of the major causes of cancer death for men. Magnetic resonance (MR) imaging is being increasingly used as an important modality to localize prostate cancer. Therefore, localizing prostate cancer in MRI with automated detection methods has become an active area of research. Many methods have been proposed for this task. However, most of previous methods focused on identifying cancer only in the peripheral zone (PZ), or classifying suspicious cancer ROIs into benign tissue and cancer tissue. Few works have been done on developing a fully automatic method for cancer localization in the entire prostate region, including central gland (CG) and transition zone (TZ). In this paper, we propose a novel learning-based multi-source integration framework to directly localize prostate cancer regions from in vivo MRI. We employ random forests to effectively integrate features from multi-source images together for cancer localization. Here, multi-source images include initially the multi-parametric MRIs (i.e., T2, DWI, and dADC) and later also the iteratively-estimated and refined tissue probability map of prostate cancer. Experimental results on 26 real patient data show that our method can accurately localize cancerous sections. The higher section-based evaluation (SBE), combined with the ROC analysis result of individual patients, shows that the proposed method is promising for in vivo MRI based prostate cancer localization, which can be used for guiding prostate biopsy, targeting the tumor in focal therapy planning, triage and follow-up of patients with active surveillance, as well as the decision making in treatment selection. The common ROC analysis with the AUC value of 0.832 and also the ROI-based ROC analysis with the AUC value of 0.883 both illustrate the effectiveness of our proposed method.

The controlled release of insulin-mimetic metal ions by the multifunction of chitosan.

Vanadium, which is an insulin-mimetic metal ion, was efficiently adsorbed on chitosan (CS). The adsorption of vanadium on CS was affected by the vanadium/CS ratio and the initial concentration of vanadium in preparative medium under constant pH condition. The vanadium-CS complex was able to control vanadium release. Moreover, a consistent control of vanadium release was achieved by incorporation of the vanadium-CS complex into a CS gel. After implantation of the CS gel retaining the vanadium-CS complex into diabetic mice, insulin-mimetic efficacy was confirmed by observation of a steady reduction in blood glucose levels. The sustained vanadium release also contributed to minimization of the side-effects. Thus, CS gel retaining the vanadium-CS complex appears promising as a vehicle for vanadium with long-term action and a low toxicity leading to its clinical use.