Image-Guided Access for Percutaneous Nephrolithotomy: A Single-Center Experience in 591 Patients.

INTRODUCTION: We present our experience in image-guided percutaneous nephrolithotomy (PCNL) access in 591 patients. MATERIALS AND METHODS: An IRB-approved review of all adult PCNL cases from 2009 to 2014 was performed. Patient data, information regarding stone size and location, procedural details, clinical success, complications by access site (upper pole versus middle or lower pole) and puncture location (supracostal versus infracostal) were recorded. RESULTS: In this study, 591 patients (314 males, 278 females, mean stone size: 23 mm, range: 4-100 mm) were included. Stone clearance was achieved in 66% of patients. There were 174 total complications (29.3%). Upper pole access was less likely to require a secondary access to achieve stone clearance (p = 0.02) and was preferentially used for both larger stones (p = 0.006) and staghorn calculi (p = 0.001). If a supracostal approach to the upper pole was used, there were significantly more complications compared to an infracostal approach (p = 0.002). CONCLUSION: Upper pole access for PCNL provides anatomic advantages for stone clearance but significantly increases the risk for complications when a supracostal puncture is required.

Management of Thyroid Nodules Seen on US Images: Deep Learning May Match Performance of Radiologists.

BackgroundManagement of thyroid nodules may be inconsistent between different observers and time consuming for radiologists. An artificial intelligence system that uses deep learning may improve radiology workflow for management of thyroid nodules.PurposeTo develop a deep learning algorithm that uses thyroid US images to decide whether a thyroid nodule should undergo a biopsy and to compare the performance of the algorithm with the performance of radiologists who adhere to American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS).Materials and MethodsIn this retrospective analysis, studies in patients referred for US with subsequent fine-needle aspiration or with surgical histologic analysis used as the standard were evaluated. The study period was from August 2006 to May 2010. A multitask deep convolutional neural network was trained to provide biopsy recommendations for thyroid nodules on the basis of two orthogonal US images as the input. In the training phase, the deep learning algorithm was first evaluated by using 10-fold cross-validation. Internal validation was then performed on an independent set of 99 consecutive nodules. The sensitivity and specificity of the algorithm were compared with a consensus of three ACR TI-RADS committee experts and nine other radiologists, all of whom interpreted thyroid US images in clinical practice.ResultsIncluded were 1377 thyroid nodules in 1230 patients with complete imaging data and conclusive cytologic or histologic diagnoses. For the 99 test nodules, the proposed deep learning algorithm achieved 13 of 15 (87%: 95% confidence interval [CI]: 67%, 100%) sensitivity, the same as expert consensus (P > .99) and higher than five of nine radiologists. The specificity of the deep learning algorithm was 44 of 84 (52%; 95% CI: 42%, 62%), which was similar to expert consensus (43 of 84; 51%; 95% CI: 41%, 62%; P = .91) and higher than seven of nine other radiologists. The mean sensitivity and specificity for the nine radiologists was 83% (95% CI: 64%, 98%) and 48% (95% CI: 37%, 59%), respectively.ConclusionSensitivity and specificity of a deep learning algorithm for thyroid nodule biopsy recommendations was similar to that of expert radiologists who used American College of Radiology Thyroid Imaging and Reporting Data System guidelines.© RSNA, 2019Online supplemental material is available for this article.

Using Artificial Intelligence to Revise ACR TI-RADS Risk Stratification of Thyroid Nodules: Diagnostic Accuracy and Utility.

Background Risk stratification systems for thyroid nodules are often complicated and affected by low specificity. Continual improvement of these systems is necessary to reduce the number of unnecessary thyroid biopsies. Purpose To use artificial intelligence (AI) to optimize the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS). Materials and Methods A total of 1425 biopsy-proven thyroid nodules from 1264 consecutive patients (1026 women; mean age, 52.9 years [range, 18-93 years]) were evaluated retrospectively. Expert readers assigned points based on five ACR TI-RADS categories (composition, echogenicity, shape, margin, echogenic foci), and a genetic AI algorithm was applied to a training set (1325 nodules). Point and pathologic data were used to create an optimized scoring system (hereafter, AI TI-RADS). Performance of the systems was compared by using a test set of the final 100 nodules with interpretations from the expert reader, eight nonexpert readers, and an expert panel. Initial performance of AI TI-RADS was calculated by using a test for differences between binomial proportions. Additional comparisons across readers were conducted by using bootstrapping; diagnostic performance was assessed by using area under the receiver operating curve. Results AI TI-RADS assigned new point values for eight ACR TI-RADS features. Six features were assigned zero points, which simplified categorization. By using expert reader data, the diagnostic performance of ACR TI-RADS and AI TI-RADS was area under the receiver operating curve of 0.91 and 0.93, respectively. For the same expert, specificity of AI TI-RADS (65%, 55 of 85) was higher (P < .001) than that of ACR TI-RADS (47%, 40 of 85). For the eight nonexpert radiologists, mean specificity for AI TI-RADS (55%) was also higher (P < .001) than that of ACR TI-RADS (48%). An interactive AI TI-RADS calculator can be viewed at http://deckard.duhs.duke.edu/∼ai-ti-rads . Conclusion An artificial intelligence-optimized Thyroid Imaging Reporting and Data System (TI-RADS) validates the American College of Radiology TI-RADS while slightly improving specificity and maintaining sensitivity. Additionally, it simplifies feature assignments, which may improve ease of use. © RSNA, 2019 Online supplemental material is available for this article.

Survival comparison of hepatocellular carcinoma patients treated with radioembolization versus nonoperative/interventional treatment.

AIM: To compare the overall survival (OS) and liver cancer-specific survival of advanced-stage hepatocellular carcinoma (HCC) patients who received transarterial radioembolization (TARE) with those who received nonoperative/interventional treatment (NOT). MATERIALS & METHODS: A total of 12,520 HCC patients from the Surveillance, Epidemiology and End Results database were categorized by treatment with either radioembolization or NOT. Kaplan-Meier and multivariate Cox regression were conducted. RESULTS: The TARE group had both a significantly longer median overall survival than the NOT group (TARE = 9 months; NOT = 2 months; p < 0.0001) and a significantly higher probability of liver cancer-specific survival (hazard ratio = 0.474). CONCLUSION: TARE appears to provide a significant survival advantage over the NOT population in advanced HCC patients.

Thermal ablation versus surgical resection for localized hepatocellular carcinoma: a population study using the SEER database.

AIM: To compare overall survival (OS) and liver cancer-specific survival (LCSS) in patients with localized hepatocellular carcinoma treated with surgical resection (SR) or thermal ablation (TA) using the Surveillance, Epidemiology and End Results database. MATERIALS & METHODS: Kaplan-Meier, competing risk and Cox regression analyses were performed after identifying patients. Propensity score matching was then applied. RESULTS: There was significantly better OS in the SR group and significantly lower probability of LCSS in the TA group. After matching, there was significantly longer OS in the SR group and a lower probability of LCSS in the TA group. CONCLUSION: SR offered a significant survival benefit over TA for localized hepatocellular carcinoma.

Differentiation of tumor vasculature heterogeneity levels in small animals based on total hemoglobin concentration using magnetic resonance-guided diffuse optical tomography in vivo.

Insight into the vasculature of the tumor in small animals has the potential to impact many areas of cancer research. The heterogeneity of the vasculature of a tumor is directly related to tumor stage and disease progression. In this small scale animal study, we investigated the feasibility of differentiating tumors with different levels of vasculature heterogeneity in vivo using a previously developed hybrid magnetic resonance imaging (MRI) and diffuse optical tomography (DOT) system for small animal imaging. Cross-sectional total hemoglobin concentration maps of 10 Fisher rats bearing R3230 breast tumors are reconstructed using multi-wavelength DOT measurements both with and without magnetic resonance (MR) structural a priori information. Simultaneously acquired MR structural images are used to guide and constrain the DOT reconstruction, while dynamic contrast-enhanced MR functional images are used as the gold standard to classify the vasculature of the tumor into two types: high versus low heterogeneity. These preliminary results show that the stand-alone DOT is unable to differentiate tumors with low and high vascular heterogeneity without structural a priori information provided by a high resolution imaging modality. The mean total hemoglobin concentrations comparing the vasculature of the tumors with low and high heterogeneity are significant (p-value 0.02) only when MR structural a priori information is utilized.

Atypical magnetic resonance imaging findings in hepatocellular carcinoma.

Magnetic resonance imaging (MRI) is currently the modality of choice to evaluate liver lesions in patients with cirrhosis and hepatitis B and C. Hepatocellular carcinoma demonstrates typical imaging findings on contrast-enhanced MRI, which are usually diagnostic. Unfortunately, a subgroup of hepatocellular carcinoma presents with atypical imaging features, and awareness of these atypical presentations is important in ensuring early diagnosis and optimal patient outcomes. Herein, we review some of the more common atypical presentations with a focus on MRI.

A True Multi-modality Approach for High Resolution Optical Imaging: Photo-Magnetic Imaging.

Multi-modality imaging leverages the competitive advantage of different imaging systems to improve the overall resolution and quantitative accuracy. Our new technique, Photo-Magnetic Imaging (PMI) is one of these true multi-modality imaging approaches, which can provide quantitative optical absorption map at MRI spatial resolution. PMI uses laser light to illuminate tissue and elevate its temperature while utilizing MR thermometry to measure the laser-induced temperature variation with high spatial resolution. The high-resolution temperature maps are later converted to tissue absorption maps by a finite element based inverse solver that is based on modeling of photon migration and heat diffusion in tissue. Previously, we have demonstrated the feasibility of PMI with phantom studies. Recently, we have managed to reduce the laser power under ANSI limit for maximum skin exposure therefore, we have well positioned PMI for in vivo imaging. Currently we are expanding our system by adding multi-wavelength imaging capability. This will allow us not only to resolve spatial distribution of tissue chromophores but also exogenous contrast agents. Although we test PMIs feasibility with animal studies, our future goal is to use PMI for breast cancer imaging due to its high translational potential.

Photo-magnetic imaging: resolving optical contrast at MRI resolution.

In this paper, we establish the mathematical framework of a novel imaging technique, namely photo-magnetic imaging (PMI). PMI uses a laser to illuminate biological tissues and measure the induced temperature variations using magnetic resonance imaging (MRI). PMI overcomes the limitation of conventional optical imaging and allows imaging of the optical contrast at MRI spatial resolution. The image reconstruction for PMI, using a finite-element-based algorithm with an iterative approach, is presented in this paper. The quantitative accuracy of PMI is investigated for various inclusion sizes, depths and absorption values. Then, a comparison between conventional diffuse optical tomography (DOT) and PMI is carried out to illustrate the superior performance of PMI. An example is presented showing that two 2 mm diameter inclusions embedded 4.5 mm deep and located side by side in a 25 mm diameter circular geometry medium are recovered as a single 6 mm diameter object with DOT. However, these two objects are not only effectively resolved with PMI, but their true concentrations are also recovered successfully.

Laser-induced photo-thermal magnetic imaging.

Due to the strong scattering nature of biological tissue, optical imaging beyond the diffusion limit suffers from low spatial resolution. In this letter, we present an imaging technique, laser-induced photo-thermal magnetic imaging (PMI), which uses laser illumination to induce temperature increase in a medium and magnetic resonance imaging to map the spatially varying temperature, which is proportional to absorbed energy. This technique can provide high-resolution images of optical absorption and can potentially be used for small animal as well as breast cancer and lymph node imaging. First, we describe the theory of PMI, including the modeling of light propagation and heat transfer in tissue. We also present experimental data with corresponding predictions from theoretical models, which show excellent agreement.

Dual-contrast dynamic MRI-DOT for small animal imaging.

In this paper we present first-of-its-kind spatially resolved enhancement kinetics of optical and magnetic resonance (MR) agents obtained by a combined MR and Diffuse Optical Tomography (MR-DOT) animal imaging system. A unique MR compatible fiber optic interface allows co-registration of MR and DOT data in space and time. High temporal resolution of the hybrid system permits acquisition of data in dynamic mode. Rats bearing a R3230 AC breast cancer tumor model are used for in vivo studies. Thirty-two optical and thirty MR images are acquired during a single imaging session that lasts nearly ten minutes. Both optical, indocyanine green (ICG), and MR contrast agents, gadolinium-DTPA (Gd-DTPA), are injected simultaneously after the acquisition of several baseline frames. Contrast enhancement time curves obtained by MR and DOT systems both indicate higher average enhancement in tumor regions, up to ten-fold for MRI and 3-fold for DOT, compared to close by non-tumor regions. This feasibility study is the first step towards clinical translation of this hybrid imaging platform. The ultimate aim is to use the enhancement kinetics of the optical agent ICG, which binds to plasma proteins, as complementary information to the kinetics of the MR agent Gd-DTPA, a small molecular agent that does not bind to plasma proteins, to better differentiate benign and malignant lesions.