Although osteoradionecrosis (ORN) is a serious complication of craniofacial radiotherapy, the current management methods remain suboptimal. Teriparatide (TPTD), a recombinant human parathyroid ...hormone (1–34), has shown beneficial effects on osseous regeneration in medication-related osteonecrosis of the jaw or periodontitis. However, TPTD therapy in irradiated bones has not been indicated yet because of the theoretical risk of osteosarcoma seen in rat models. Hence, we first report here two patients with tongue cancer with late-emerging ORN who were successfully treated with TPTD for 4–6 months with serum calcium and vitamin D supplementation. In contrast to the usual progress of ORN, the bone defect regenerated well and bone turnover markers including serum C-terminal telopeptide of type 1 collagen and osteocalcin were restored with TPTD therapy. Our experience might suggest that TPTD therapy with careful monitoring can provide an effective treatment option for patients with ORN in select refractory cases, with the benefits outweighing the potential risks.
Summary Adult articular chondrocytes undergo slow senescence and dedifferentiation during in vitro expansion, restricting successful cartilage regeneration. A complete understanding of the molecular ...signaling pathways involved in the senescence and dedifferentiation of chondrocytes is essential in order to better characterize chondrocytes for cartilage tissue engineering applications. During expansion, cell fate is determined by the change in expression of various genes in response to aspects of the microenvironment, including oxidative stress, mechanical stress, and unsuitable culture conditions. Rapid senescence or dedifferentiation not only results in the loss of the chondrocytic phenotype but also enhances production of inflammatory mediators and matrix-degrading enzymes. This review focuses on the two groups of genes that play direct and indirect roles in the induction of senescence and dedifferentiation. Numerous degenerative signaling pathways associated with these genes have been reported. Upregulation of the genes interleukin 1 beta ( IL-1β ), p53 , p16 , p21 , and p38 mitogen-activated protein kinase ( MAPK ) is responsible for the direct induction of senescence, whereas downregulation of the genes transforming growth factor-beta ( TGF-β ), bone morphogenetic protein-2 ( BMP-2 ), SRY (sex determining region Y)-box 9 ( SOX9 ), and insulin-like growth factor-1 ( IGF-1 ), indirectly induces senescence. In senescent and dedifferentiated chondrocytes, it was found that TGF - β , BMP - 2 , SOX9 , and IGF-1 are downregulated, while the levels of IL - 1β , p53 , p16 , p21 , and p38 MAPK are upregulated followed by inhibition of the normal molecular functioning of the chondrocytes. This review helps to elucidate the underlying mechanism in degenerative cartilage disease, which may help to improve cartilage tissue regeneration techniques.
The goals of this review paper on deep learning (DL) in medical imaging and radiation therapy are to (a) summarize what has been achieved to date; (b) identify common and unique challenges, and ...strategies that researchers have taken to address these challenges; and (c) identify some of the promising avenues for the future both in terms of applications as well as technical innovations. We introduce the general principles of DL and convolutional neural networks, survey five major areas of application of DL in medical imaging and radiation therapy, identify common themes, discuss methods for dataset expansion, and conclude by summarizing lessons learned, remaining challenges, and future directions.
Immune checkpoint inhibitors (ICIs) have been shown to be beneficial for some patients with advanced non-small-cell lung cancer (NSCLC). However, the underlying mechanisms mediating the limited ...response to ICIs remain unclear.
We carried out whole-exome sequencing on 198 advanced NSCLC tumors that had been sampled before anti-programmed cell death 1 (anti-PD-1)/programmed death-ligand 1 (PD-L1) therapy. Detailed clinical characteristics were collected on these patients. We designed a new method to estimate human leukocyte antigen (HLA)-corrected tumor mutation burden (TMB), a modification which considers the loss of heterozygosity of HLA from conventional TMB. We carried out external validation of our findings utilizing 89 NSCLC samples and 110 melanoma samples from two independent cohorts of immunotherapy-treated patients.
Homology-dependent recombination deficiency was identified in 37 patients (18.7%) and was associated with longer progression-free survival (PFS; P = 0.049). Using the HLA-corrected TMB, non-responders to ICIs were identified, despite having a high TMB (top 25%). Ten patients (21.3% of the high TMB group) were reclassified from the high TMB group into the low TMB group. The objective response rate (ORR), PFS, and overall survival (OS) were all lower in these patients compared with those of the high TMB group (ORR: 20% versus 59%, P = 0.0363; PFS: hazard ratio = 2.91, P = 0.007; OS: hazard ratio = 3.43, P = 0.004). Multivariate analyses showed that high HLA-corrected TMB was associated with a significant survival advantage (hazard ratio = 0.44, P = 0.015), whereas high conventional TMB was not associated with a survival advantage (hazard ratio = 0.63, P = 0.118). Applying this approach to the independent cohorts of 89 NSCLC patients and 110 melanoma patients, TMB-based survival prediction was significantly improved.
HLA-corrected TMB can reconcile the observed disparity in relationships between TMB and ICI responses, and is of predictive and prognostic value for ICI therapies.
•TMB alone is not sufficiently reliable or accurate as a biomarker of response to ICIs in NSCLC.•TMB-based survival prediction is improved by using the HLA-corrected TMB algorithm (TMB in combination with loss of heterozygosity of HLA).•Notably, additional predictive and prognostic value of the HLA-corrected TMB is not limited to certain types of cancer.•The HLA-corrected TMB could be a new strategy for selecting patients who may benefit from immunotherapy.
In this paper, we developed a deep convolutional neural network (CNN) for the classification of malignant and benign masses in digital breast tomosynthesis (DBT) using a multi-stage transfer learning ...approach that utilized data from similar auxiliary domains for intermediate-stage fine-tuning. Breast imaging data from DBT, digitized screen-film mammography, and digital mammography totaling 4039 unique regions of interest (1797 malignant and 2242 benign) were collected. Using cross validation, we selected the best transfer network from six transfer networks by varying the level up to which the convolutional layers were frozen. In a single-stage transfer learning approach, knowledge from CNN trained on the ImageNet data was fine-tuned directly with the DBT data. In a multi-stage transfer learning approach, knowledge learned from ImageNet was first fine-tuned with the mammography data and then fine-tuned with the DBT data. Two transfer networks were compared for the second-stage transfer learning by freezing most of the CNN structures versus freezing only the first convolutional layer. We studied the dependence of the classification performance on training sample size for various transfer learning and fine-tuning schemes by varying the training data from 1% to 100% of the available sets. The area under the receiver operating characteristic curve (AUC) was used as a performance measure. The view-based AUC on the test set for single-stage transfer learning was 0.85 ± 0.05 and improved significantly (p <; 0.05 ) to 0.91 ± 0.03 for multi-stage learning. This paper demonstrated that, when the training sample size from the target domain is limited, an additional stage of transfer learning using data from a similar auxiliary domain is advantageous.
We fabricated Pt-functionalised hydrogen gas sensors on AlGaN/GaN heterojunction platform and investigated the influence of GaN-cap layer on the sensing characteristics. Pt-Schottky diodes with ...GaN-cap layer exhibited a larger change of Schottky barrier height than ones with no GaN-cap layer when hydrogen gas was detected. Technology computer-aided design simulation indicated that the increase of electron concentration at heterojunction can be magnified by a larger change of barrier height. The AlGaN/GaN FET-type sensors with Pt catalyst on the gate area demonstrated significant enhancement of hydrogen gas sensitivity from 16 to 35% at 200°C when GaN cap layer was employed.
Purpose:
The authors are developing a computerized system for bladder segmentation in CT urography (CTU) as a critical component for computer-aided detection of bladder cancer.
Methods:
A ...deep-learning convolutional neural network (DL-CNN) was trained to distinguish between the inside and the outside of the bladder using 160 000 regions of interest (ROI) from CTU images. The trained DL-CNN was used to estimate the likelihood of an ROI being inside the bladder for ROIs centered at each voxel in a CTU case, resulting in a likelihood map. Thresholding and hole-filling were applied to the map to generate the initial contour for the bladder, which was then refined by 3D and 2D level sets. The segmentation performance was evaluated using 173 cases: 81 cases in the training set (42 lesions, 21 wall thickenings, and 18 normal bladders) and 92 cases in the test set (43 lesions, 36 wall thickenings, and 13 normal bladders). The computerized segmentation accuracy using the DL likelihood map was compared to that using a likelihood map generated by Haar features and a random forest classifier, and that using our previous conjoint level set analysis and segmentation system (CLASS) without using a likelihood map. All methods were evaluated relative to the 3D hand-segmented reference contours.
Results:
With DL-CNN-based likelihood map and level sets, the average volume intersection ratio, average percent volume error, average absolute volume error, average minimum distance, and the Jaccard index for the test set were 81.9% ± 12.1%, 10.2% ± 16.2%, 14.0% ± 13.0%, 3.6 ± 2.0 mm, and 76.2% ± 11.8%, respectively. With the Haar-feature-based likelihood map and level sets, the corresponding values were 74.3% ± 12.7%, 13.0% ± 22.3%, 20.5% ± 15.7%, 5.7 ± 2.6 mm, and 66.7% ± 12.6%, respectively. With our previous CLASS with local contour refinement (LCR) method, the corresponding values were 78.0% ± 14.7%, 16.5% ± 16.8%, 18.2% ± 15.0%, 3.8 ± 2.3 mm, and 73.9% ± 13.5%, respectively.
Conclusions:
The authors demonstrated that the DL-CNN can overcome the strong boundary between two regions that have large difference in gray levels and provides a seamless mask to guide level set segmentation, which has been a problem for many gradient-based segmentation methods. Compared to our previous CLASS with LCR method, which required two user inputs to initialize the segmentation, DL-CNN with level sets achieved better segmentation performance while using a single user input. Compared to the Haar-feature-based likelihood map, the DL-CNN-based likelihood map could guide the level sets to achieve better segmentation. The results demonstrate the feasibility of our new approach of using DL-CNN in combination with level sets for segmentation of the bladder.
Transfer learning in deep convolutional neural networks (DCNNs) is an important step in its application to medical imaging tasks. We propose a multi-task transfer learning DCNN with the aim of ...translating the 'knowledge' learned from non-medical images to medical diagnostic tasks through supervised training and increasing the generalization capabilities of DCNNs by simultaneously learning auxiliary tasks. We studied this approach in an important application: classification of malignant and benign breast masses. With Institutional Review Board (IRB) approval, digitized screen-film mammograms (SFMs) and digital mammograms (DMs) were collected from our patient files and additional SFMs were obtained from the Digital Database for Screening Mammography. The data set consisted of 2242 views with 2454 masses (1057 malignant, 1397 benign). In single-task transfer learning, the DCNN was trained and tested on SFMs. In multi-task transfer learning, SFMs and DMs were used to train the DCNN, which was then tested on SFMs. N-fold cross-validation with the training set was used for training and parameter optimization. On the independent test set, the multi-task transfer learning DCNN was found to have significantly (p = 0.007) higher performance compared to the single-task transfer learning DCNN. This study demonstrates that multi-task transfer learning may be an effective approach for training DCNN in medical imaging applications when training samples from a single modality are limited.
Cross-sectional X-ray imaging has become the standard for staging most solid organ malignancies. However, for some malignancies such as urinary bladder cancer, the ability to accurately assess local ...extent of the disease and understand response to systemic chemotherapy is limited with current imaging approaches. In this study, we explored the feasibility that radiomics-based predictive models using pre- and post-treatment computed tomography (CT) images might be able to distinguish between bladder cancers with and without complete chemotherapy responses. We assessed three unique radiomics-based predictive models, each of which employed different fundamental design principles ranging from a pattern recognition method via deep-learning convolution neural network (DL-CNN), to a more deterministic radiomics feature-based approach and then a bridging method between the two, utilizing a system which extracts radiomics features from the image patterns. Our study indicates that the computerized assessment using radiomics information from the pre- and post-treatment CT of bladder cancer patients has the potential to assist in assessment of treatment response.
A highly integrated bidirectional analogue front-end (AFE) circuit for interfacing capacitive micromachined ultrasound transducer in medical imaging systems is presented. The proposed AFE features a ...reconfigurable high-voltage (HV) pulser which generates over 15 V pulses at 2.6 MHz in the transmit mode and also operates as an HV isolation switch for the receive path circuits, thereby saving significant silicon area. The AFE also includes a low-power low-noise preamplifier which achieves 103 dBΩ transimpedance gain at 2.6 and 5 MHz bandwidth while operating at 1.65 V supply voltage. The AFE integrated circuit (IC) is implemented using 0.18 µm standard CMOS process and the total area of the single-channel core is 0.052 mm2 which is more than 65% of size reduction in comparison to previous work.