Multiparametric Ultrasound Imaging of Prostate Cancer Using Deep Neural Networks.

OBJECTIVE: A deep neural network (DNN) was trained to generate a multiparametric ultrasound (mpUS) volume from four input ultrasound-based modalities (acoustic radiation force impulse [ARFI] imaging, shear wave elasticity imaging [SWEI], quantitative ultrasound-midband fit [QUS-MF], and B-mode) for the detection of prostate cancer. METHODS: A DNN was trained using co-registered ARFI, SWEI, MF, and B-mode data obtained in men with biopsy-confirmed prostate cancer prior to radical prostatectomy (15 subjects, comprising 980,620 voxels). Data were obtained using a commercial scanner that was modified to allow user control of the acoustic beam sequences and provide access to the raw image data. For each subject, the index lesion and a non-cancerous region were manually segmented using visual confirmation based on whole-mount histopathology data. RESULTS: In a prostate phantom, the DNN increased lesion contrast-to-noise ratio (CNR) compared to a previous approach that used a linear support vector machine (SVM). In the in vivo test datasets (n = 15), the DNN-based mpUS volumes clearly portrayed histopathology-confirmed prostate cancer and significantly improved CNR compared to the linear SVM (2.79 ± 0.88 vs. 1.98 ± 0.73, paired-sample t-test p < 0.001). In a sub-analysis in which the input modalities to the DNN were selectively omitted, the CNR decreased with fewer inputs; both stiffness- and echogenicity-based modalities were important contributors to the multiparametric model. CONCLUSION: The findings from this study indicate that a DNN can be optimized to generate mpUS prostate volumes with high CNR from ARFI, SWEI, MF, and B-mode and that this approach outperforms a linear SVM approach.

Combined ARFI and Shear Wave Imaging of Prostate Cancer: Optimizing Beam Sequences and Parameter Reconstruction Approaches.

This study demonstrates the implementation of a shear wave reconstruction algorithm that enables concurrent acoustic radiation force impulse (ARFI) imaging and shear wave elasticity imaging (SWEI) of prostate cancer and zonal anatomy. The combined ARFI/SWEI sequence uses closely spaced push beams across the lateral field of view and simultaneously tracks both on-axis (within the region of excitation) and off-axis (laterally offset from the excitation) after each push beam. Using a large number of push beams across the lateral field of view enables the collection of higher signal-to-noise ratio (SNR) shear wave data to reconstruct the SWEI volume than is typically acquired. The shear wave arrival times were determined with cross-correlation of shear wave velocity signals in two dimensions after 3-D directional filtering to remove reflection artifacts. To combine data from serially interrogated lateral push locations, arrival times from different pushes were aligned by estimating the shear wave propagation time between push locations. Shear wave data acquired in an elasticity lesion phantom and reconstructed using this algorithm demonstrate benefits to contrast-to-noise ratio (CNR) with increased push beam density and 3-D directional filtering. Increasing the push beam spacing from 0.3 to 11.6 mm (typical for commercial SWEI systems) resulted in a 53% decrease in CNR. In human in vivo data, this imaging approach enabled high CNR (1.61-1.86) imaging of histologically-confirmed prostate cancer. The in vivo images had improved spatial resolution and CNR and fewer reflection artifacts as a result of the high push beam density, the high shear wave SNR, the use of multidimensional directional filtering, and the combination of shear wave data from different push beams.

Prostate Cancer Detection Using 3-D Shear Wave Elasticity Imaging.

Transrectal ultrasound (TRUS) B-mode imaging provides insufficient sensitivity and specificity for prostate cancer (PCa) targeting when used for biopsy guidance. Shear wave elasticity imaging (SWEI) is an elasticity imaging technique that has been commercially implemented and is sensitive and specific for PCa. We have developed a SWEI system capable of 3-D data acquisition using a dense acoustic radiation force (ARF) push approach that leads to enhanced shear wave signal-to-noise ratio compared with that of the commercially available SWEI systems and facilitates screening of the entire gland before biopsy. Additionally, we imaged and assessed 36 patients undergoing radical prostatectomy using 3-D SWEI and determined a shear wave speed threshold separating PCa from healthy prostate tissue with sensitivities and specificities akin to those for multiparametric magnetic resonance imaging fusion biopsy. The approach measured the mean shear wave speed in each prostate region to be 4.8 m/s (Young's modulus E = 69.1 kPa) in the peripheral zone, 5.3 m/s (E = 84.3 kPa) in the central gland and 6.0 m/s (E = 108.0 kPa) for PCa with statistically significant (p < 0.0001) differences among all regions. Three-dimensional SWEI receiver operating characteristic analyses identified a threshold of 5.6 m/s (E = 94.1 kPa) to separate PCa from healthy tissue with a sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and area under the curve (AUC) of 81%, 82%, 69%, 89% and 0.84, respectively. Additionally, a shear wave speed ratio was assessed to normalize for tissue compression and patient variability, which yielded a threshold of 1.11 to separate PCa from healthy prostate tissue and was accompanied by a substantial increase in specificity, PPV and AUC, where the sensitivity, specificity, PPV, NPV and AUC were 75%, 90%, 79%, 88% and 0.90, respectively. This work illustrates the feasibility of using 3-D SWEI data to detect and localize PCa and demonstrates the benefits of normalizing for applied compression during data acquisition for use in biopsy targeting studies.

Deep Convolutional Neural Networks for Displacement Estimation in ARFI Imaging.

Ultrasound elasticity imaging in soft tissue with acoustic radiation force requires the estimation of displacements, typically on the order of several microns, from serially acquired raw data A-lines. In this work, we implement a fully convolutional neural network (CNN) for ultrasound displacement estimation. We present a novel method for generating ultrasound training data, in which synthetic 3-D displacement volumes with a combination of randomly seeded ellipsoids are created and used to displace scatterers, from which simulated ultrasonic imaging is performed using Field II. Network performance was tested on these virtual displacement volumes, as well as an experimental ARFI phantom data set and a human in vivo prostate ARFI data set. In the simulated data, the proposed neural network performed comparably to Loupas's algorithm, a conventional phase-based displacement estimation algorithm; the rms error was [Formula: see text] for the CNN and 0.73 [Formula: see text] for Loupas. Similarly, in the phantom data, the contrast-to-noise ratio (CNR) of a stiff inclusion was 2.27 for the CNN-estimated image and 2.21 for the Loupas-estimated image. Applying the trained network to in vivo data enabled the visualization of prostate cancer and prostate anatomy. The proposed training method provided 26 000 training cases, which allowed robust network training. The CNN had a computation time that was comparable to Loupas's algorithm; further refinements to the network architecture may provide an improvement in the computation time. We conclude that deep neural network-based displacement estimation from ultrasonic data is feasible, providing comparable performance with respect to both accuracy and speed compared to current standard time-delay estimation approaches.

Multiparametric Ultrasound for Targeting Prostate Cancer: Combining ARFI, SWEI, QUS and B-Mode.

Diagnosing prostate cancer through standard transrectal ultrasound (TRUS)-guided biopsy is challenging because of the sensitivity and specificity limitations of B-mode imaging. We used a linear support vector machine (SVM) to combine standard TRUS imaging data with acoustic radiation force impulse (ARFI) imaging data, shear wave elasticity imaging (SWEI) data and quantitative ultrasound (QUS) midband fit data to enhance lesion contrast into a synthesized multiparametric ultrasound volume. This SVM was trained and validated using a subset of 20 patients and tested on a second subset of 10 patients. Multiparametric US led to a statistically significant improvements in contrast, contrast-to-noise ratio (CNR) and generalized CNR (gCNR) when compared with standard TRUS B-mode and SWEI; in contrast and CNR when compared with MF; and in CNR when compared with ARFI. ARFI, MF and SWEI also outperformed TRUS B-mode in contrast, with MF outperforming B-mode in CNR and gCNR as well. ARFI, although only yielding statistically significant differences in contrast compared with TRUS B-mode, captured critical qualitative features for lesion identification. Multiparametric US enhanced lesion visibility metrics and is a promising technique for targeted TRUS-guided prostate biopsy in the future.

Right-Sided Versus Left-Sided Pneumonectomy After Induction Therapy for Non-Small Cell Lung Cancer.

BACKGROUND: A right-sided pneumonectomy after induction therapy for non-small cell lung cancer (NSCLC) has been shown to be associated with significant perioperative risk. We examined the effect of laterality on long-term survival after induction therapy and pneumonectomy using the National Cancer Data Base. METHODS: Perioperative and long-term outcomes of patients who underwent pneumonectomy after induction chemotherapy, with or without radiotherapy, from 2004 to 2014 in the National Cancer Data Base were evaluated using multivariable Cox proportional hazards modeling and propensity score-matched analysis. RESULTS: During the study period, 1,465 patients (right, 693 [47.3%]; left, 772 [52.7%]) met inclusion criteria. Right-sided pneumonectomy was associated with significantly higher 30-day (8.2% [57 of 693] vs 4.2% [32 of 772], p < 0.01) and 90-day mortality (13.6% [94 of 693] vs 7.9% [61 of 772], p < 0.01), and right-sided pneumonectomy was a predictor of higher 90-day mortality (odds ratio, 2.23; p < 0.01). However, overall 5-year survival between right and left pneumonectomy was not significantly different in unadjusted (37.6% [95% confidence interval {CI}, 0.34 to 0.42] vs 35% [95% CI, 0.32 to 0.39], log-rank p = 0.94) or multivariable analysis (hazard ratio, 1.07; 95% CI, 0.92 to 1.25; p = 0.40). A propensity score-matched analysis of 810 patients found no significant differences in 5-year survival between the right-sided versus left-sided groups (34.7% [95% CI, 0.30 to 0.40] vs 34.1%, [95% CI, 0.29 to 0.39], log-rank p = 0.86). CONCLUSIONS: In this national analysis, right-sided pneumonectomy after induction therapy was associated with a significantly higher perioperative but not worse long-term mortality compared to a left-sided procedure.

Long-term Survival After Surgery Compared With Concurrent Chemoradiation for Node-negative Small Cell Lung Cancer.

OBJECTIVE: To determine whether surgery with adjuvant chemotherapy offers a survival advantage over concurrent chemoradiation for patients with cT1-2N0M0 small cell lung cancer (SCLC). BACKGROUND: Although surgery with adjuvant chemotherapy is the recommended treatment for patients with cT1-2N0M0 SCLC per international guidelines, there have been no prospective or retrospective studies evaluating the impact of surgery versus optimal medical management for cT1-2N0M0 SCLC. METHODS: Outcomes of patients with cT1-2N0M0 SCLC who underwent surgery with adjuvant chemotherapy or concurrent chemoradiation in the National Cancer Data Base (2003-2011) were evaluated using Cox proportional hazards analyses and propensity-score-matched analyses. RESULTS: During the study period, 681 (30%) patients underwent surgery with adjuvant chemotherapy and 1620 (70%) underwent concurrent chemoradiation. After propensity-score matching, all 14 covariates were well balanced between the surgery (n = 501) and concurrent chemoradiation (n = 501) groups. Surgery was associated with a higher overall survival (OS) than concurrent chemoradiation (5-year OS 47.6% vs 29.8%, P < 0.01). To minimize selection bias due to comorbidities, we limited the propensity-matched analysis to 492 patients with no comorbidities; surgery remained associated with a higher OS than concurrent chemoradiation (5-year OS 49.2% vs 32.5%, P < 0.01). CONCLUSIONS: In a national analysis, surgery with adjuvant chemotherapy was used in the minority of patients for early stage SCLC. Surgery with adjuvant chemotherapy for node-negative SCLC was associated with improved long-term survival when compared to concurrent chemoradiation. These results suggest a significant underuse of surgery among patients with early stage SCLC and support an increased role of surgery in multimodality therapy for cT1-2N0M0 SCLC.

Self-determined shapes and velocities of giant near-zero drag gas cavities.

Minimizing the retarding force on a solid moving in liquid is the canonical problem in the quest for energy saving by friction and drag reduction. For an ideal object that cannot sustain any shear stress on its surface, theory predicts that drag force will fall to zero as its speed becomes large. However, experimental verification of this prediction has been challenging. We report the construction of a class of self-determined streamlined structures with this free-slip surface, made up of a teardrop-shaped giant gas cavity that completely encloses a metal sphere. This stable gas cavity is formed around the sphere as it plunges at a sufficiently high speed into the liquid in a deep tank, provided that the sphere is either heated initially to above the Leidenfrost temperature of the liquid or rendered superhydrophobic in water at room temperature. These sphere-in-cavity structures have residual drag coefficients that are typically less than [Formula: see text] those of solid objects of the same dimensions, which indicates that they experienced very small drag forces. The self-determined shapes of the gas cavities are shown to be consistent with the Bernoulli equation of potential flow applied on the cavity surface. The cavity fall velocity is not arbitrary but is uniquely predicted by the sphere density and cavity volume, so larger cavities have higher characteristic velocities.

Surgery Versus Optimal Medical Management for N1 Small Cell Lung Cancer.

BACKGROUND: Adjuvant chemotherapy has been demonstrated to improve the outcomes of patients with N1 non-small cell lung cancer. It is unknown whether patients previously thought to have unresectable small cell lung cancer (SCLC) may have tumors amenable to surgery if adjuvant therapies can be given. This study was undertaken to evaluate whether surgery, in the setting of modern adjuvant therapies, can be beneficial for patients with N1-positive SCLC. METHODS: Patients with clinical T1-3 N1 M0 SCLC who underwent concurrent chemoradiation versus surgery and adjuvant therapy in the National Cancer Data Base from 2003 to 2011 were examined. Overall survival was assessed using Kaplan-Meier and Cox proportional hazards analysis and propensity score-matched analysis. RESULTS: Of 1,041 patients with cT1-3 N1 M0 SCLC who met inclusion criteria, 96 patients (9%) underwent surgery and adjuvant chemotherapy with or without radiation and 945 (91%) underwent concurrent chemoradiation alone. Multivariable Cox modeling demonstrated that surgery with adjuvant chemotherapy with or without radiation (hazard ratio 0.74, 95% confidence interval: 0.56 to 0.97) was associated with improved survival compared with concurrent chemoradiation. After propensity matching, surgery with adjuvant chemotherapy with or without radiation was associated with improved 5-year survival compared with concurrent chemoradiation (31.4% versus 26.3%). CONCLUSIONS: In an analysis of a national population-based cancer database, surgery followed by adjuvant chemotherapy with or without radiation for cT1-3 N1 SCLC had improved outcomes compared with concurrent chemoradiation. These results support the re-evaluation of the role of surgery in multimodality therapy for N1 SCLC in a clinical trial setting.

Role of Adjuvant Therapy in a Population-Based Cohort of Patients With Early-Stage Small-Cell Lung Cancer.

PURPOSE: Data on optimal adjuvant therapy after complete resection of small-cell lung cancer (SCLC) are limited, and in particular, there have been no studies evaluating the role of adjuvant chemotherapy, with or without prophylactic cranial irradiation, relative to no adjuvant therapy for stage T1-2N0M0 SCLC. This National Cancer Data Base analysis was performed to determine the potential benefits of adjuvant chemotherapy with and without prophylactic cranial irradiation in patients who undergo complete resection for early-stage small-cell lung cancer. PATIENTS AND METHODS: Overall survival of patients with pathologic T1-2N0M0 SCLC who underwent complete resection in the National Cancer Data Base from 2003 to 2011, stratified by adjuvant therapy regimen, was evaluated using Kaplan-Meier and Cox proportional hazards analysis. Patients treated with induction therapy and those who died within 30 days of surgery were excluded from analysis. RESULTS: Of 1,574 patients who had pT1-2N0M0 SCLC during the study period, 954 patients (61%) underwent complete R0 resection with a 5-year survival of 47%. Adjuvant therapy was administered to 59% of patients (n = 566), including chemotherapy alone (n = 354), chemoradiation (n = 190, including 99 patients who underwent cranial irradiation), and radiation alone (n = 22). Compared with surgery alone, adjuvant chemotherapy with or without radiation was associated with significantly improved survival. In addition, multivariable Cox modeling demonstrated that treatment with adjuvant chemotherapy (hazard ratio [HR], 0.78; 95% CI, 0.63 to 0.95) or chemotherapy with radiation directed at the brain (HR, 0.52; 95% CI, 0.36 to 0.75) was associated with improved survival when compared with no adjuvant therapy. CONCLUSION: Patients with pT1-2N0M0 SCLC treated with surgical resection alone have worse outcomes than those who undergo resection with adjuvant chemotherapy alone or chemotherapy with cranial irradiation.

Study of fluid and transport properties of porous anodic aluminum membranes by dynamic atomic force microscopy.

Recent work on carbon nanotubes (CNT) has focused on their potential application in water treatment as a result of their predicted and observed enhanced flow rates. Recent work on the lesser-known porous anodic alumina membranes (PAAMs) has also shown flow enhancement, albeit at only a fraction of what has been observed in CNTs. Despite their potential applications, little research has been conducted on PAAMs' hydrodynamic properties, and in this Article we present experimental results and theoretical models that explore the fluid flow behavior around and through these membranes. The experiments were conducted using an atomic force microscope (AFM) that pushed a solid silica particle against PAAMs that were characterized with different pore diameters. Furthermore, the PAAMs were classified as either closed or open, with the latter allowing fluid to pass through. The theoretical model developed to describe the experimental data incorporates Derjaguin-Landau-Verwey-Overbeek (DLVO) effects, cantilever drag, and hydrodynamic forces. By using the slip boundary condition for the hydrodynamic forces, we were able to fit the model to experimental findings and also demonstrate that the difference between closed and open PAAMs was negligible. The slip lengths did not correspond to any physical feature of the PAAMs, but our model does provide a simple yet effective means of describing the hydrodynamics for not only PAAMs but for membranes in general.

Layer with reduced viscosity at water-oil interfaces probed by fluorescence correlation spectroscopy.

The two-dimensional diffusion of isolated molecular tracers at the water-n-alkane interface was studied with fluorescence correlation spectroscopy. The interfacial diffusion coefficients of larger tracers with a hydrodynamic radius of 4.0 nm agreed well with the values calculated from the macroscopic viscosities of the two bulk phases. However, for small molecule tracers with hydrodynamic radii of only 1.0 and 0.6 nm, notable deviations were observed, indicating the existence of an interfacial region with reduced effective viscosity and increased mobility.

Anomalous pH dependent stability behavior of surfactant-free nonpolar oil drops in aqueous electrolyte solutions.

Recent advances in atomic force microscopy (AFM) force measurement techniques have allowed the direct measurement and theoretical interpretation of the interaction between a liquid droplet and a solid surface or between two liquid droplets. In this study, we investigated the interaction across an aqueous thin film between fluorocarbon (perfluoropentane) droplets, hydrocarbon (tetradecane) droplets, and a droplet and a flat mica surface in the absence of stabilizers. It was found that even at a relatively elevated electrolyte concentration of 0.1 M NaNO3, depending on the solution pH, interactions between two identical droplets or a droplet and a mica surface could be repulsive. A simple theoretical analysis of the magnitude and range of these interactive forces suggests that the DLVO theory cannot explain the observed behavior. The measured force behavior is discussed in the context of ion adsorption, and the arising charging effects, at the bare oil-water interface.