Prospective prediction of the major component of urinary stone composition with dual-source dual-energy CT in vivo

Prediction of the Uric Acid Component in Nephrolithiasis Using Simple Clinical Information about Metabolic Disorder and Obesity: A Machine Learning-Based Model

There is a great need for a diagnostic tool using simple clinical information collected from patients to diagnose uric acid (UA) stones in nephrolithiasis. We built a predictive model making use of machine learning (ML) methodologies entering simple parameters easily obtained at the initial clinical visit. Socio-demographic, health, and clinical data from two cohorts (A and B), both diagnosed with nephrolithiasis, one between 2012 and 2016 and the other between June and December 2020, were collected before nephrolithiasis treatment. A ML-based model for predicting UA stones in nephrolithiasis was developed using eight simple parameters-sex, age, gout, diabetes mellitus, body mass index, estimated glomerular filtration rate, bacteriuria, and urine pH. Data from Cohort A were used for model training and validation (ratio 3:2), while data from Cohort B were used only for validation. One hundred and forty-six (13.3%) out of 1098 patients in Cohort A and 3 (4.23%) out of 71 patients in Cohort B had pure UA stones. For Cohort A, our model achieved a validation AUC (area under ROC curve) of 0.842, with 0.8475 sensitivity and 0.748 specificity. For Cohort B, our model achieved 0.936 AUC, with 1.0 sensitivity, and 0.912 specificity. This ML-based model provides a convenient and reliable method for diagnosing urolithiasis. Using only eight readily available clinical parameters, including information about metabolic disorder and obesity, it distinguished pure uric acid stones from other stones before treatment.

Value of artificial intelligence model based on unenhanced computed tomography of urinary tract for preoperative prediction of calcium oxalate monohydrate stones in vivo

Background

Urolithiasis is a global disease with a high incidence and recurrence rate, and stone composition is closely related to the choice of treatment and preventive measures. Calcium oxalate monohydrate (COM) is the most common in clinical practice, which is hard and difficult to fragment. Preoperative identification of its components and selection of effective surgical methods can reduce the risk of patients having a second operation. Methods that can be used for stone composition analysis include infrared spectroscopy, X-ray diffraction, and polarized light microscopy, but they are all performed on stone specimens in vitro after surgery. This study aimed to design and develop an artificial intelligence (AI) model based on unenhanced computed tomography (CT) images of the urinary tract, and to investigate the predictive ability of the model for COM stones in vivo preoperatively, so as to provide surgeons with more accurate diagnostic information.

Methods

Preoperative unenhanced CT images of patients with urinary calculi whose components were determined by infrared spectroscopy in a single center were retrospectively analyzed, including 337 cases of COM stones and 170 of non-COM stones. All images were manually segmented and the image features were extracted, and randomly divided into the training and testing sets in a ratio of 7:3. The least absolute shrinkage and selection operation algorithm (LASSO) was used to construct the AI model, and classification of the training and testing sets was carried out.

Results

A total of 1,218 radiomics imaging features were extracted, and 8 features with non-zero coefficients were finally obtained. The sensitivity, specificity and accuracy of the AI model were 90.5%, 84.3% and 88.5% for the training set, and 90.1%, 84.3% and 88.3% for the testing set. The area under the curve was 0.935 for the training set and 0.933 for the testing set.

Conclusions

The AI model based on unenhanced CT images of the urinary tract can predict COM and non-COM stones in vivo preoperatively, and the model has high sensitivity, specificity and accuracy.

A multicenter study to develop a non-invasive radiomic model to identify urinary infection stone in vivo using machine-learning

Urolithiasis is a common urological disease, and treatment strategy options vary between different stone types. However, accurate detection of stone composition can only be performed in vitro. The management of infection stones is particularly challenging with yet no effective approach to pre-operatively identify infection stones from non-infection stones. Therefore, we aimed to develop a radiomic model for preoperatively identifying infection stones with multicenter validation. In total, 1198 eligible patients with urolithiasis from three centers were divided into a training set, an internal validation set, and two external validation sets. Stone composition was determined by Fourier transform infrared spectroscopy. A total of 1316 radiomic features were extracted from the pre-treatment Computer Tomography images of each patient. Using the least absolute shrinkage and selection operator algorithm, we identified a radiomic signature that achieved favorable discrimination in the training set, which was confirmed in the validation sets. Moreover, we then developed a radiomic model incorporating the radiomic signature, urease-producing bacteria in urine, and urine pH based on multivariate logistic regression analysis. The nomogram showed favorable calibration and discrimination in the training and three validation sets (area under the curve [95% confidence interval], 0.898 [0.840-0.956], 0.832 [0.742-0.923], 0.825 [0.783-0.866], and 0.812 [0.710-0.914], respectively). Decision curve analysis demonstrated the clinical utility of the radiomic model. Thus, our proposed radiomic model can serve as a non-invasive tool to identify urinary infection stones in vivo, which may optimize disease management in urolithiasis and improve patient prognosis.

Single-energy CT predicts uric acid stones with accuracy comparable to dual-energy CT-prospective validation of a quantitative method

OBJECTIVES

To prospectively validate three quantitative single-energy CT (SE-CT) methods for classifying uric acid (UA) and non-uric acid (non-UA) stones.

METHODS

Between September 2018 and September 2019, 116 study participants were prospectively included in the study if they had at least one 3-20-mm urinary stone on an initial urinary tract SE-CT scan. An additional dual-energy CT (DE-CT) scan was performed, limited to the stone of interest. Additionally, to include a sufficient number of UA stones, eight participants with confirmed UA stone on DE-CT were retrospectively included. The SE-CT stone features used in the prediction models were (1) maximum attenuation (maxHU) and (2) the peak point Laplacian (ppLapl) calculated at the position in the stone with maxHU. Two prediction models were previously published methods (ppLapl-maxHU and maxHU) and the third was derived from the previous results based on the k-nearest neighbors (kNN) algorithm (kNN-ppLapl-maxHU). The three methods were evaluated on this new independent stone dataset. The reference standard was the CT vendor's DE-CT application for kidney stones.

RESULTS

Altogether 124 participants (59 ± 14 years, 91 men) with 106 non-UA and 37 UA stones were evaluated. For classification of UA and non-UA stones, the sensitivity, specificity, and accuracy were 100% (37/37), 97% (103/106), and 98% (140/143), respectively, for kNN-ppLapl-maxHU; 95% (35/37), 98% (104/106), and 97% (139/143) for ppLapl-maxHU; and 92% (34/37), 94% (100/106), and 94% (134/143) for maxHU.

CONCLUSION

A quantitative SE-CT method (kNN-ppLapl-maxHU) can classify UA stones with accuracy comparable to DE-CT.

KEY POINTS

• Single-energy CT is the first-line diagnostic tool for suspected renal colic. • A single-energy CT method based on the internal urinary stone attenuation distribution can classify urinary stones into uric acid and non-uric acid stones with high accuracy. • This immensely increases the availability of in vivo stone analysis.

Dual-Energy CT for Urinary Stone Evaluation

PURPOSE OF REVIEW

Conventional CT imaging is an excellent tool for the diagnosis of nephrolithiasis however is limited in its ability to detect stone composition. Dual-energy CT (DECT) scans have demonstrated promise in overcoming this limitation. We review the current utility of DECT in nephrolithiasis.

RECENT FINDINGS

DECT is superior to conventional CT in differentiating uric acid stones from non-uric acid stones, with numerous studies reporting sensitivities and specificities approaching > 95%. Dose reduction protocols incorporating low-dose CT scans are commonly used, providing significantly lower effective radiation doses compared to conventional CT. DECT remains an effective diagnostic tool in patients with large body habitus. DECT can accurately detect uric acid stones, which can help guide which stones may be suitable to medical dissolution. Further studies evaluating the effectiveness of DECT in guiding management of patients with nephrolithiasis can help to promote its widespread use.

Efficacy of single-source rapid kV-switching dual-energy CT for characterization of non-uric acid renal stones: a prospective ex vivo study using anthropomorphic phantom

PURPOSE

To investigate the accuracy of rapid kV-switching single-source dual-energy computed tomography (rsDECT) for prediction of classes of non-uric-acid stones.

MATERIALS AND METHODS

Non-uric-acid renal stones retrieved via percutaneous nephrolithotomy were prospectively collected between January 2017 and February 2018 in a single institution. Only stones ≥ 5 mm and with pure composition (i.e., ≥ 80% composed of one component) were included. Stone composition was determined using Fourier Transform Infrared Spectroscopy. The stones were scanned in 32-cm-wide anthropomorphic whole-body phantom using rsDECT. The effective atomic number (Zeff), the attenuation at 40 keV (HU40), 70 keV (HU70), and 140 keV (HU140) virtual monochromatic sets of images as well as the ratios between the attenuations were calculated. Values of stone classes were compared using ANOVA and Mann-Whitney U test. Receiver operating curves and area under curve (AUC) were calculated. A p value < 0.05 was considered statistically significant.

RESULTS

The final study sample included 31 stones from 31 patients consisting of 25 (81%) calcium-based, 4 (13%) cystine, and 2 (6%) struvite pure stones. The mean size of the stones was 9.9 ± 2.4 mm. The mean Zeff of the stones was 12.01 ± 0.54 for calcium-based, 11.10 ± 0.68 for struvite, and 10.23 ± 0.75 for cystine stones (p < 0.001). Zeff had the best efficacy to separate different classes of stones. The calculated AUC was 0.947 for Zeff; 0.833 for HU40; 0.880 for HU70; and 0.893 for HU140.

CONCLUSION

Zeff derived from rsDECT has superior performance to HU and attenuation ratios for separation of different classes of non-uric-acid stones.

Aktualisierung der S2k-Leitlinie zur Diagnostik, Therapie und Metaphylaxe der Urolithiasis (AWMF Registernummer 043-025)

Die Zunahme des medizinischen Wissens, technische Neuerungen gemeinsam mit demographischem Wandel stellen eine Herausforderung an die Neukonzeption von Leitlinien und klinischen Studien dar. Die vorliegende S2k-Leitlinie, die sich ausschließlich mit Nieren- und Harnleitersteinen beschäftigt, soll die Behandlung von Harnsteinpatienten in Klinik und Praxis unterstützen, aber auch Patienteninformationen zur Urolithiasis geben. Die zunehmende interdisziplinäre Zusammenarbeit in der Steintherapie zeigt sich auch an der Anzahl beteiligter Fachgruppen und Arbeitsgemeinschaften in der Erstellung des neuen Leitlinienupdates. Die vorliegende, aus einem interdisziplinären Konsensusprozess hervorgegangene S2k-Leitlinie stellt die aktuellen Empfehlungen praxisnah dar und gibt Entscheidungshilfen für Diagnostik‑, Therapie- und Metaphylaxemaßnahmen auf Basis von Expertenmeinungen und verfügbaren Evidenzgrundlagen aus der Literatur.

Optimized management of urolithiasis by coloured stent-stone contrast using dual-energy computed tomography (DECT)

BACKGROUND

We analysed in vitro the appearance of commonly used ureteral stents with dual-energy computed tomography (DECT) and we used these characteristics to optimize the differentiation between stents and adjacent stone.

METHODS

We analysed in vitro a selection of 36 different stents from 7 manufacturers. They were placed in a self-build phantom model and measured using the SOMATOM® Force Dual Source CT-Scanner (Siemens, Forchheim, Germany). Each sample was scanned at various tube potentials of 80 and 150 peak kilovoltage (kVp), 90 and 150 kVp and 100 and 150 kVp. The syngo Post-Processing Suite software program (Siemens, Forchheim, Germany) was used for differentiation based on a 3-material decomposition algorithm (UA, calcium, urine) according to our standard stone protocol.

RESULTS

Stents composed of polyurethane appeared blue and silicon-based stents were red on the image. The determined appearances were constant for various peak kilovoltage (kVp) values. The coloured stent-stone-contrast displayed on DECT improves monitoring, especially of small calculi adjacent to indwelling ureteral stents.

CONCLUSION

Both urinary calculi and ureteral stents can be accurately differentiated by a distinct appearance on DECT. For the management of urolithiasis patients can be monitored more easily and accurately using DECT if the stent shows a different colour than the adjacent stone.

Uric acid versus non-uric acid urinary stones: differentiation with single energy CT texture analysis

AIM

To evaluate the accuracy of computed tomography (CT) texture analysis (TA) to differentiate uric acid (UA) stones from non-UA stones on unenhanced CT in patients with urinary calculi with ex vivo Fourier transform infrared spectroscopy (FTIR) as the reference standard.

MATERIALS AND METHODS

Fourteen patients with 18 UA stones and 31 patients with 32 non-UA stones were included. All the patients had preoperative CT evaluation and subsequent surgical removal of the stones. CTTA was performed on CT images using commercially available research software. Each texture feature was evaluated using the non-parametric Mann-Whitney test. Receiver operating characteristic (ROC) curves were created and the area under the ROC curve (AUC) was calculated for texture parameters that were significantly different. The features were used to train support vector machine (SVM) classifiers. Diagnostic accuracy was evaluated.

RESULTS

Compared to non-UA stones, UA stones had significantly lower mean, standard deviation and mean of positive pixels but higher kurtosis (p<0.001) on both unfiltered and filtered texture scales. There were no significant differences in entropy or skewness between UA and non-UA stones. The average SVM accuracy of texture features for differentiating UA from non-UA stones ranged from 88% to 92% (after 10-fold cross validation). A model incorporating standard deviation, skewness, and kurtosis from unfiltered texture scale images resulted in an AUC of 0.965±00.029 with a sensitivity of 94.4% and specificity of 93.7%.

CONCLUSION

CTTA can be used to accurately differentiate UA stones from non-UA stones in vivo using unenhanced CT images.