Outcomes of Artificial Intelligence Volumetric Assessment of Kidneys and Renal Tumors for Preoperative Assessment of Nephron Sparing Interventions

Artificial Intelligence for Improved Hepatosplenomegaly Diagnosis

Hepatosplenomegaly is commonly diagnosed by radiologists based on single dimension measurements and heuristic cut-offs. Volumetric measurements may be more accurate for diagnosing organ enlargement. Artificial intelligence techniques may be able to automatically calculate liver and spleen volume and facilitate more accurate diagnosis. After IRB approval, 2 convolutional neural networks (CNN) were developed to automatically segment the liver and spleen on a training dataset comprised of 500 single-phase, contrast-enhanced CT abdomen and pelvis examinations. A separate dataset of ten thousand sequential examinations at a single institution was segmented with these CNNs. Performance was evaluated on a 1% subset and compared with manual segmentations using Sorensen-Dice coefficients and Pearson correlation coefficients. Radiologist reports were reviewed for diagnosis of hepatomegaly and splenomegaly and compared with calculated volumes. Abnormal enlargement was defined as greater than 2 standard deviations above the mean. Median Dice coefficients for liver and spleen segmentation were 0.988 and 0.981, respectively. Pearson correlation coefficients of CNN-derived estimates of organ volume against the gold-standard manual annotation were 0.999 for the liver and spleen (P < 0.001). Average liver volume was 1556.8 ± 498.7 cc and average spleen volume was 194.6 ± 123.0 cc. There were significant differences in average liver and spleen volumes between male and female patients. Thus, the volume thresholds for ground-truth determination of hepatomegaly and splenomegaly were determined separately for each sex. Radiologist classification of hepatomegaly was 65% sensitive, 91% specific, with a positive predictive value (PPV) of 23% and an negative predictive value (NPV) of 98%. Radiologist classification of splenomegaly was 68% sensitive, 97% specific, with a positive predictive value (PPV) of 50% and a negative predictive value (NPV) of 99%. Convolutional neural networks can accurately segment the liver and spleen and may be helpful to improve radiologist accuracy in the diagnosis of hepatomegaly and splenomegaly.

The national utilization of nonoperative management for small renal masses over 10 years

BACKGROUND

Management of small renal masses often involves a nonoperative approach, but there is a paucity of information about the use and associated predictors of such approaches. This study aimed to determine the trends in and predictors of use of nonoperative management of small renal masses.

METHODS

Using data from the National Cancer Database for localized small renal masses (N0/M0, cT1a) diagnosed between 2010 and 2020, we conducted a cross-sectional study. Nonoperative management was defined as expectant management (active surveillance or watchful waiting) or focal ablation. Adjusted odds ratios (AORs) were calculated using multivariable logistic regression models.

RESULTS

Of the 156 734 patients included, 10.5% underwent expectant management, and 13.9% underwent focal ablation. Later year of diagnosis was associated with a higher likelihood of nonoperative management. In 2020, the odds of receiving expectant management and focal ablation were 90% (AOR = 1.90, 95% confidence interval [CI] = 1.71 to 2.11) and 44% (AOR = 1.44, 95% CI = 1.31 to 1.57) higher, respectively, than in 2010. Black patients had increased odds of expectant management (AOR = 1.47, 95% CI = 1.39 to 1.55) but decreased odds of focal ablation (AOR = 0.93, 95% CI = 0.88 to 0.99).

CONCLUSION

Over the decade, the use nonoperative management of small renal masses increased, with expectant management more frequently used than focal ablation among Black patients. Possible explanations include race-based differences in physicians' risk assessments and resource allocation. Adjusting for Black race in calculations for glomerular filtration rate could influence the differential uptake of these techniques through deflated glomerular filtration rate calculations. These findings highlight the need for research and policies to ensure equitable use of less invasive treatments in small renal masses.

Artificial Intelligence in Urooncology: What We Have and What We Expect

INTRODUCTION

Artificial intelligence is transforming healthcare by driving innovation, automation, and optimization across various fields of medicine. The aim of this study was to determine whether artificial intelligence (AI) techniques can be used in the diagnosis, treatment planning, and monitoring of urological cancers.

METHODOLOGY

We conducted a thorough search for original and review articles published until 31 May 2022 in the PUBMED/Scopus database. Our search included several terms related to AI and urooncology. Articles were selected with the consensus of all authors.

RESULTS

Several types of AI can be used in the medical field. The most common forms of AI are machine learning (ML), deep learning (DL), neural networks (NNs), natural language processing (NLP) systems, and computer vision. AI can improve various domains related to the management of urologic cancers, such as imaging, grading, and nodal staging. AI can also help identify appropriate diagnoses, treatment options, and even biomarkers. In the majority of these instances, AI is as accurate as or sometimes even superior to medical doctors.

CONCLUSIONS

AI techniques have the potential to revolutionize the diagnosis, treatment, and monitoring of urologic cancers. The use of AI in urooncology care is expected to increase in the future, leading to improved patient outcomes and better overall management of these tumors.

Volumetric Analysis of Renal Masses as Predictors of Partial Nephrectomy Outcomes

Objective: To examine the role of endophytic tumor volume (TV) assessment (endophycity) on perioperative partial nephrectomy (PN) outcomes. Patients and Methods: Retrospective review of 212 consecutive laparoscopic and open partial nephrectomies from single institution using preoperative imaging and 1-year follow-up. Demographics, comorbidities, RENAL nephrometry scores, and all peri- and postoperative outcomes were recorded. Volumetric analysis performed using imaging software, independently assessed by two blinded radiologists. Univariate and multivariate statistical analysis were completed to assess predictive value of endophycity for all clinically meaningful outcomes. Results: Among those undergoing minimally invasive surgery (MIS), lower tumor endophycity was associated with higher likelihood of trifecta outcome (negative surgical margin, <10% decline in estimated glomerular filtration rate, the absence of complications) irrespective of max tumor size. For MIS, estimated blood loss increased with greater tumor endophycity regardless of tumor size. Among those who underwent open partial nephrectomy, lower tumor endophycity was associated with trifecta outcomes for tumors >4 cm only. On multivariate analysis with log-scaled odds ratios (OR), tumor endophycity and total kidney volume had the strongest correlation with tumor-related complications (OR = 3.23, 2.66). The analysis identified that tumor endophycity and TV on imaging were inversely correlated with of trifecta outcomes (OR = 0.53 for both covariates). Conclusions: Volumetric assessment of tumor endophycity performed well in identifying PN outcomes. As automated imaging software improves, volumetric analysis may prove to be a useful adjunct in preoperative planning and patient counseling.

Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms

Purpose

To automatically evaluate renal masses in CT images by using a cascade 3D U-Net- and ResNet-based method to accurately segment and classify focal renal lesions.

Material and Methods

We used an institutional dataset comprising 610 CT image series from 490 patients from August 2009 to August 2021 to train and evaluate the proposed method. We first determined the boundaries of the kidneys on the CT images utilizing a 3D U-Net-based method to be used as a region of interest to search for renal mass. An ensemble learning model based on 3D U-Net was then used to detect and segment the masses, followed by a ResNet algorithm for classification. Our algorithm was evaluated with an external validation dataset and kidney tumor segmentation (KiTS21) challenge dataset.

Results

The algorithm achieved a Dice similarity coefficient (DSC) of 0.99 for bilateral kidney boundary segmentation in the test set. The average DSC for renal mass delineation using the 3D U-Net was 0.75 and 0.83. Our method detected renal masses with recalls of 84.54% and 75.90%. The classification accuracy in the test set was 86.05% for masses (<5 mm) and 91.97% for masses (≥5 mm).

Conclusion

We developed a deep learning-based method for fully automated segmentation and classification of renal masses in CT images. Testing of this algorithm showed that it has the capability of accurately localizing and classifying renal masses.

Nephrometric scoring system: Recent advances and outlooks

A nephrometry scoring system is a key standard to evaluate the feasibility of partial nephrectomy (PN). Whether based on two-dimensional or three-dimensional images, simplicity, effectiveness, and practicality are the keys to the nephrometric scoring system. Since the emergence of RENAL score in 2009, numerous scoring systems based on different anatomical parameters are established to seek accurately and few parameters to assess the risk of PN and complications. This study aimed to achieve a three-game winning streak in PN more easily and efficiently (negative resection margin, maximum preservation of normal nephron function, and avoiding short-term and long-term complications). Using PubMed, we counted 28 kinds of nephrometric scoring systems. We considered only English literatures published and excluded editorials, commentaries, and meeting abstracts. To the best of our knowledge, this is to date and most comprehensive summary as well as an outlook of the nephrometric scoring system.

Artificial intelligence and radiomics in evaluation of kidney lesions: a comprehensive literature review

Radiomics and artificial intelligence (AI) may increase the differentiation of benign from malignant kidney lesions, differentiation of angiomyolipoma (AML) from renal cell carcinoma (RCC), differentiation of oncocytoma from RCC, differentiation of different subtypes of RCC, to predict Fuhrman grade, to predict gene mutation through molecular biomarkers and to predict treatment response in metastatic RCC undergoing immunotherapy. Neural networks analyze imaging data. Statistical, geometrical, textural features derived are giving quantitative data of contour, internal heterogeneity and gray zone features of lesions. A comprehensive literature review was performed, until July 2022. Studies investigating the diagnostic value of radiomics in differentiation of renal lesions, grade prediction, gene alterations, molecular biomarkers and ongoing clinical trials have been analyzed. The application of AI and radiomics could lead to improved sensitivity, specificity, accuracy in detecting and differentiating between renal lesions. Standardization of scanner protocols will improve preoperative differentiation between benign, low-risk cancers and clinically significant renal cancers and holds the premises to enhance the diagnostic ability of imaging tools to characterize renal lesions.

Artificial intelligence in oncologic imaging

Radiology is integral to cancer care. Compared to molecular assays, imaging has its advantages. Imaging as a noninvasive tool can assess the entirety of tumor unbiased by sampling error and is routinely acquired at multiple time points in oncological practice. Imaging data can be digitally post-processed for quantitative assessment. The ever-increasing application of Artificial intelligence (AI) to clinical imaging is challenging radiology to become a discipline with competence in data science, which plays an important role in modern oncology. Beyond streamlining certain clinical tasks, the power of AI lies in its ability to reveal previously undetected or even imperceptible radiographic patterns that may be difficult to ascertain by the human sensory system. Here, we provide a narrative review of the emerging AI applications relevant to the oncological imaging spectrum and elaborate on emerging paradigms and opportunities. We envision that these technical advances will change radiology in the coming years, leading to the optimization of imaging acquisition and discovery of clinically relevant biomarkers for cancer diagnosis, staging, and treatment monitoring. Together, they pave the road for future clinical translation in precision oncology.

Robotic partial nephrectomy in 3D virtual reconstructions era: is the paradigm changed?

CONTEXT

The development of a tailored, patient-specific medical and surgical approach is becoming object of intense research. In kidney oncologic surgery, where a clear understanding of case-specific surgical anatomy is considered a key point to optimize the perioperative outcomes, such philosophy gained increasing importance. Recently, important advances in 3D virtual modeling technologies have fueled the interest for their application in the field of robotic minimally invasive surgery for kidney tumors.

OBJECTIVE

To provide a synthesis of current applications of 3D virtual models for robot-assisted partial nephrectomy.

EVIDENCE ACQUISITION

Medline, PubMed, the Cochrane Database, and Embase were screened for Literature regarding the use of 3D virtual models for robot-assisted partial nephrectomy (RAPN).

EVIDENCE SYNTHESIS

The use of 3D virtual models for RAPN has been tested in different settings, including surgical indication and planning, intraoperative guidance, and training. Currently, several studies are available on the application of this technology for surgical planning, demonstrating impact on clinical outcomes such as renal function recovery, whilst experiences concerning their intraoperative application for navigation are still experimental. One of the latest innovations in this field is represented by the development of dedicated softwares able to automatically overlap the 3D virtual models to the real anatomy, to perform augmented reality procedures.

CONCLUSIONS

The available Literature suggests a potentially crucial role of 3D virtual reconstructions during RAPN. Encouraging results concerning surgical planning and indication, intraoperative navigation, and surgical training are available. In the future, artificial intelligence may represent the key to further improve the 3D virtual modeling technology during RAPN.