A Deep Learning Method of Human Identification from Radar Signal for Daily Sleep Health Monitoring.

Radar signal has been shown as a promising source for human identification. In daily home sleep-monitoring scenarios, large-scale motion features may not always be practical, and the heart motion or respiration data may not be as ideal as they are in a controlled laboratory setting. Human identification from radar sequences is still a challenging task. Furthermore, there is a need to address the open-set recognition problem for radar sequences, which has not been sufficiently studied. In this paper, we propose a deep learning-based approach for human identification using radar sequences captured during sleep in a daily home-monitoring setup. To enhance robustness, we preprocess the sequences to mitigate environmental interference before employing a deep convolution neural network for human identification. We introduce a Principal Component Space feature representation to detect unknown sequences. Our method is rigorously evaluated using both a public data set and a set of experimentally acquired radar sequences. We report a labeling accuracy of 98.2% and 96.8% on average for the two data sets, respectively, which outperforms the state-of-the-art techniques. Our method excels at accurately distinguishing unknown sequences from labeled ones, with nearly 100% detection of unknown samples and minimal misclassification of labeled samples as unknown.

Monte Carlo simulation of physical dose enhancement in core-shell magnetic gold nanoparticles with TOPAS.

The application of metal nanoparticles (MNPs) as sensitization materials is a common strategy that is used to study dose enhancement in radiotherapy. Recent in vitro tests have revealed that magnetic gold nanoparticles (NPs) can be used in cancer therapy under a magnetic field to enhance the synergistic efficiency in radiotherapy and photothermal therapy. However, magnetic gold NPs have rarely been studied as sensitization materials. In this study, we obtained further results of the sensitization properties of the magnetic gold NPs (Fe3O4@AuNPs) with or without magnetic field using the TOPAS-nBio Monte Carlo (MC) toolkit. We analyzed the properties of Fe3O4@AuNP in a single NP model and in a cell model under monoenergetic photons and brachytherapy, and we investigated whether the magnetic field contributes to the physical sensitization process. Our results revealed that the dose enhancement factor (DEF) of Fe3O4@AuNPs was lower than that of gold nanoparticles (AuNPs) in a single NP and in a cell irradiated by monoenergetic photons. But it's worth mentioning that under a magnetic field, the DEF of targeted Fe3O4@AuNPs in a cell model with a clinical brachytherapy source was 22.17% (cytoplasm) and 6.89% (nucleus) higher than those of AuNPs (50 mg/mL). The Fe3O4@AuNPs were proved as an effective sensitization materials when combined with the magnetic field in MC simulation for the first time, which contributes to the research on in vitro tests on radiosensitization as well as clinical research in future.

Monte Carlo study of TG-43 dosimetry parameters of GammaMed Plus high dose rate 192 Ir brachytherapy source using TOPAS.

PURPOSE: To develop a simulation model for GammaMed Plus high dose rate 192 Ir brachytherapy source in TOPAS Monte Carlo software and validate it by calculating the TG-43 dosimetry parameters and comparing them with published data. METHODS: We built a model for GammaMed Plus high dose rate brachytherapy source in TOPAS. The TG-43 dosimetry parameters including air-kerma strength SK , dose-rate constant Λ, radial dose function gL (r), and 2D anisotropy function F(r,θ) were calculated using Monte Carlo simulation with Geant4 physics models and NNDC 192 Ir spectrum. Calculations using an old 192 Ir spectrum were also carried out to evaluate the impact of incident spectrum and cross sections. The results were compared with published data. RESULTS: For calculations using the NNDC spectrum, the air-kerma strength per unit source activity SK /A and Λ were 1.0139 × 10-7 U/Bq and 1.1101 cGy.h-1 .U-1 , which were 3.56% higher and 0.62% lower than the reference values, respectively. The gL (r) agreed with reference values within 1% for radial distances from 2 mm to 20 cm. For radial distances of 1, 3, 5, and 10 cm, the agreements between F(r,θ) from this work and the reference data were within 1.5% for 15° < Î¸ < 165°, and within 4% for all θ values. The discrepancies were attributed to the updated source spectrum and cross sections. They caused deviations of the SK /A of 2.90% and 0.64%, respectively. As for gL (r), they caused average deviations of -0.22% and 0.48%, respectively. Their impact on F(r,θ) was not quantified for the relatively high statistical uncertainties, but basically they did not result in significant discrepancies. CONCLUSION: A model for GammaMed Plus high dose rate 192 Ir brachytherapy source was developed in TOPAS and validated following TG-43 protocols, which can be used for future studies. The impact of updated incident spectrum and cross sections on the dosimetry parameters was quantified.

Radiographical assessment of tumour stroma and treatment outcomes using deep learning: a retrospective, multicohort study.

BACKGROUND: The tumour stroma microenvironment plays an important part in disease progression and its composition can influence treatment response and outcomes. Histological evaluation of tumour stroma is limited by access to tissue, spatial heterogeneity, and temporal evolution. We aimed to develop a radiological signature for non-invasive assessment of tumour stroma and treatment outcomes. METHODS: In this multicentre, retrospective study, we analysed CT images and outcome data of 2209 patients with resected gastric cancer from five independent cohorts recruited from two centres (Nanfang Hospital of Southern Medical University [Guangzhou, China] and Sun Yat-sen University Cancer Center [Guangzhou, China]). Patients with histologically confirmed gastric cancer, at least 15 lymph nodes harvested, preoperative abdominal CT available, and complete clinicopathological and follow-up data were eligible for inclusion. Tumour tissue was collected for patients in the training cohort (321 patients), internal validation cohort one (246 patients), and external validation cohort one (128 patients). Four stroma classes were defined according to the protein expression of α-smooth muscle actin and periostin assessed by immunohistochemistry. The primary objective was to predict the histologically based stroma classes by using preoperative CT images. We trained a deep convolutional neural network model using the training cohort and tested the model in the internal and external validation cohort one. We evaluated the model's association with prognosis in the training cohort, two internal, and two external validation cohorts and compared outcomes of patients who received or did not receive adjuvant chemotherapy. FINDINGS: The deep-learning model achieved a high diagnostic accuracy for assessing tumour stroma in both internal validation cohort one (area under the receiver operating characteristic curve [AUC] 0·96-0·98]) and external validation cohort one (AUC 0·89-0·94). The stromal imaging signature was significantly associated with disease-free survival and overall survival in all cohorts (p<0·0001). The predicted stroma classes remained an independent prognostic factor adjusting for clinicopathological variables including tumour size, stage, differentiation, and Lauren histology. In patients with stage II or III disease in predicted stroma classes one and two subgroups, patients who received adjuvant chemotherapy had improved survival compared with those who did not (in those with stage II disease hazard ratio [HR] 0·48 [95% CI 0·29-0·77], p=0·0021; and in those with stage III disease HR 0·70 [0·57-0·85], p=0·00042). However, in the other two subgroups adjuvant chemotherapy was not associated with survival and might even be detrimental in the predicted stroma class 4 subgroup (HR 1·48 [1·08-2·03], p=0·013). INTERPRETATION: The deep-learning model could allow for accurate and non-invasive evaluation of tumour stroma from CT images in gastric cancer. The radiographical model predicted chemotherapy outcomes and could be used in combination with clinicopathological criteria to refine prognosis and inform treatment decisions of patients with gastric cancer. FUNDING: None.

Monte Carlo simulations of energy deposition and DNA damage using TOPAS-nBio.

PURPOSE: Monte Carlo (MC) track structure codes are commonly used for predicting energy deposition and radiation-induced DNA damage at the nanometer scale. Various simulation parameters such as physics model, DNA model, and direct damage threshold have been developed. The differences in adopted parameters lead to disparity in calculation results, which requires quantitative evaluation. METHODS: Three simulation configurations were implemented in TOPAS-nBio MC toolkit to investigate the impact of physics models, DNA model, and direct damage threshold on the prediction of energy deposition and DNA damage. Dose point kernels (DPKs) of electrons and nanometer-sized volumes irradiated with electrons, protons, and alpha particles were utilized to evaluate the impact of physics models on energy deposition. Proton irradiation of plasmid DNA was used to investigate the disparity in single-strand break and double-strand break (DSB) yields caused by differences in physics models, DNA models, and direct damage thresholds. RESULTS: Electron DPKs obtained with different physics models show similar trends but different diffusiveness and maximums. Energy deposition distributions in nanometer-sized volumes irradiated with electrons, protons, and alpha particles calculated using different physics models have the same trend although discrepancies can be observed at the lowest and highest energy deposits. Strand breaks from incident protons in DNA plasmids vary with adopted parameters. For the configurations in this study, changing physics model, DNA model, and direct damage threshold can cause differences of up to 57%, 69%, and 15% in DSB yields, respectively. All these simulation results are essentially in agreement with previously published simulation or experimental studies. CONCLUSION: All the physics models, DNA models, and direct damage thresholds investigated in this study are applicable to predict energy deposition and DNA damage. Although the choice of parameters can lead to disparity in simulation results, which serves as a reference for future studies.

A consistency evaluation of signal-to-noise ratio in the quality assessment of human brain magnetic resonance images.

BACKGROUND: Quality assessment of medical images is highly related to the quality assurance, image interpretation and decision making. As to magnetic resonance (MR) images, signal-to-noise ratio (SNR) is routinely used as a quality indicator, while little knowledge is known of its consistency regarding different observers. METHODS: In total, 192, 88, 76 and 55 brain images are acquired using T2*, T1, T2 and contrast-enhanced T1 (T1C) weighted MR imaging sequences, respectively. To each imaging protocol, the consistency of SNR measurement is verified between and within two observers, and white matter (WM) and cerebral spinal fluid (CSF) are alternately used as the tissue region of interest (TOI) for SNR measurement. The procedure is repeated on another day within 30 days. At first, overlapped voxels in TOIs are quantified with Dice index. Then, test-retest reliability is assessed in terms of intra-class correlation coefficient (ICC). After that, four models (BIQI, BLIINDS-II, BRISQUE and NIQE) primarily used for the quality assessment of natural images are borrowed to predict the quality of MR images. And in the end, the correlation between SNR values and predicted results is analyzed. RESULTS: To the same TOI in each MR imaging sequence, less than 6% voxels are overlapped between manual delineations. In the quality estimation of MR images, statistical analysis indicates no significant difference between observers (Wilcoxon rank sum test, p w ≥ 0.11; paired-sample t test, p p ≥ 0.26), and good to very good intra- and inter-observer reliability are found (ICC, p icc ≥ 0.74). Furthermore, Pearson correlation coefficient (r p ) suggests that SNRwm correlates strongly with BIQI, BLIINDS-II and BRISQUE in T2* (r p ≥ 0.78), BRISQUE and NIQE in T1 (r p ≥ 0.77), BLIINDS-II in T2 (r p ≥ 0.68) and BRISQUE and NIQE in T1C (r p ≥ 0.62) weighted MR images, while SNRcsf correlates strongly with BLIINDS-II in T2* (r p ≥ 0.63) and in T2 (r p ≥ 0.64) weighted MR images. CONCLUSIONS: The consistency of SNR measurement is validated regarding various observers and MR imaging protocols. When SNR measurement performs as the quality indicator of MR images, BRISQUE and BLIINDS-II can be conditionally used for the automated quality estimation of human brain MR images.

Rotator cuff tendon assessment using magic-angle insensitive 3D ultrashort echo time cones magnetization transfer (UTE-Cones-MT) imaging and modeling with histological correlation.

BACKGROUND: Rotator cuff tendons (RCTs) are challenging to image due to the "magic angle effect" and their short T2 . PURPOSE: To assess the degree of magic angle sensitivity of human RCTs and to utilize a 3D ultrashort echo time Cones sequence with magnetization transfer preparation (UTE-Cones-MT) and two-pool quantitative MT modeling with histological correlation. We hypothesized that MT parameters would be less sensitive to the magic angle compared with conventional T2 measurements. STUDY TYPE: Prospective imaging pathologic correlation. SPECIMEN: Twenty cadaveric rotator cuff tendons were imaged at five sample orientations ranging from 0-90° relative to the B0 field. FIELD STRENGTH/SEQUENCE: 3T/3D UTE-Cones-MT and Carr-Purcell-Meiboom-Gill (CPMG). ASSESSMENT: Two-pool quantitative MT modeling parameters and T2 values were calculated in regions of interest drawn by a medical physicist. Histopathological analysis was performed and mild and severe tendinopathy groups were assigned by a histopathologist and histotechnician. STATISTICAL TESTS: Coefficients of variations (CVs) were calculated for measures between the different orientations and group means were compared for each measure. RESULTS: CVs of T2 and macromolecular fractions between orientations were 26.14 ± 16.82% and 6.18 ± 2.77% (mean ± SD), respectively. T2 measurements at 0°, 27°, 70°, and 90° showed significant differences between the two histological groups (P = 0.004, 0.008, 0.003, and 0.015, respectively), but not at 55° (P = 0.611). Mean T2 value ranges between orientations for the mild and severe tendinopathy groups were 15.27-30.32 msec and 20.81-35.85 msec, respectively, showing overlap despite statistically significant differences (P = 0.003). Macromolecular fractions at all angles showed significant differences between the two groups (P < 0.0001). Mean fraction ranges between orientations for the mild and severe tendinopathy groups were 14.32-17.17% and 10.00-13.75% respectively (P < 0.0001) with no overlap. DATA CONCLUSION: Compared with T2 , macromolecular fraction obtained with the 3D UTE-Cones-MT technique is resistant to the magic angle effect and is more sensitive to RCT degeneration. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2017.

A shallow convolutional neural network for blind image sharpness assessment.

Blind image quality assessment can be modeled as feature extraction followed by score prediction. It necessitates considerable expertise and efforts to handcraft features for optimal representation of perceptual image quality. This paper addresses blind image sharpness assessment by using a shallow convolutional neural network (CNN). The network takes single feature layer to unearth intrinsic features for image sharpness representation and utilizes multilayer perceptron (MLP) to rate image quality. Different from traditional methods, CNN integrates feature extraction and score prediction into an optimization procedure and retrieves features automatically from raw images. Moreover, its prediction performance can be enhanced by replacing MLP with general regression neural network (GRNN) and support vector regression (SVR). Experiments on Gaussian blur images from LIVE-II, CSIQ, TID2008 and TID2013 demonstrate that CNN features with SVR achieves the best overall performance, indicating high correlation with human subjective judgment.

A block-wise approximate parallel implementation for ART algorithm on CUDA-enabled GPU.

Computed tomography (CT) has been widely used to acquire volumetric anatomical information in the diagnosis and treatment of illnesses in many clinics. However, the ART algorithm for reconstruction from under-sampled and noisy projection is still time-consuming. It is the goal of our work to improve a block-wise approximate parallel implementation for the ART algorithm on CUDA-enabled GPU to make the ART algorithm applicable to the clinical environment. The resulting method has several compelling features: (1) the rays are allotted into blocks, making the rays in the same block parallel; (2) GPU implementation caters to the actual industrial and medical application demand. We test the algorithm on a digital shepp-logan phantom, and the results indicate that our method is more efficient than the existing CPU implementation. The high computation efficiency achieved in our algorithm makes it possible for clinicians to obtain real-time 3D images.

Segmentation of brain magnetic resonance angiography images based on MAP-MRF with multi-pattern neighborhood system and approximation of regularization coefficient.

Existing maximum a posteriori probability and Markov random field (MRF) models have limitations associated with: (1) the ordinary neighborhood system being unable to differentiate subtle changes due to several-to-one correspondence within the neighborhood; and (2) difficulty finding an appropriate parameter to balance between the spatial context and the data likelihood. Aiming at overcoming the limitations and applications to segmentation of cerebral vessels from magnetic resonance angiography images, we have proposed (1) a multi-pattern neighborhood system and corresponding energy equation to enable the MRF model for segmenting fine cerebral vessels with complicated context; and (2) an iterative approximation algorithm based on the maximum pseudo-likelihood and the space coding mode for the automatic parameter estimation of high level model of MRF. In the implementation, two computational strategies have been employed to speed up: the candidate space of cerebral vessels based on a threshold of the response to multi-scale filtering, and parallel computation of major equations. Three phantoms simulating segmentation challenges of vessels have been devised to quantitatively validate the algorithm. In addition, 10 three-dimensional clinical data sets have been used to validate the algorithm qualitatively. It has been shown that the proposed method could yield smaller error, improve the spatial resolution of MRF model, and better balance the smoothing and data likelihood than the traditional trial-and-error estimation. Comparative studies have shown that the proposed method is better than the 3 segmentation algorithms (Hassouna et al., 2006; Hao et al., 2008; Gao et al., 2011) in terms of segmentation accuracy, robustness to noise and varying curvatures as well as radii.

Objective assessment of deformable image registration in radiotherapy: a multi-institution study.

The looming potential of deformable alignment tools to play an integral role in adaptive radiotherapy suggests a need for objective assessment of these complex algorithms. Previous studies in this area are based on the ability of alignment to reproduce analytically generated deformations applied to sample image data, or use of contours or bifurcations as ground truth for evaluation of alignment accuracy. In this study, a deformable phantom was embedded with 48 small plastic markers, placed in regions varying from high contrast to roughly uniform regional intensity, and small to large regional discontinuities in movement. CT volumes of this phantom were acquired at different deformation states. After manual localization of marker coordinates, images were edited to remove the markers. The resulting image volumes were sent to five collaborating institutions, each of which has developed previously published deformable alignment tools routinely in use. Alignments were done, and applied to the list of reference coordinates at the inhale state. The transformed coordinates were compared to the actual marker locations at exhale. A total of eight alignment techniques were tested from the six institutions. All algorithms performed generally well, as compared to previous publications. Average errors in predicted location ranged from 1.5 to 3.9 mm, depending on technique. No algorithm was uniformly accurate across all regions of the phantom, with maximum errors ranging from 5.1 to 15.4 mm. Larger errors were seen in regions near significant shape changes, as well as areas with uniform contrast but large local motion discontinuity. Although reasonable accuracy was achieved overall, the variation of error in different regions suggests caution in globally accepting the results from deformable alignment.