Errors in the ultrasound diagnosis of the kidneys, ureters and urinary bladder

Current progress in artificial intelligence-assisted medical image analysis for chronic kidney disease: A literature review

Chronic kidney disease (CKD) causes irreversible damage to kidney structure and function. Arising from various etiologies, risk factors for CKD include hypertension and diabetes. With a progressively increasing global prevalence, CKD is an important public health problem worldwide. Medical imaging has become an important diagnostic tool for CKD through the non-invasive identification of macroscopic renal structural abnormalities. Artificial intelligence (AI)-assisted medical imaging techniques aid clinicians in the analysis of characteristics that cannot be easily discriminated by the naked eye, providing valuable information for the identification and management of CKD. Recent studies have demonstrated the effectiveness of AI-assisted medical image analysis as a clinical support tool using radiomics- and deep learning-based AI algorithms for improving the early detection, pathological assessment, and prognostic evaluation of various forms of CKD, including autosomal dominant polycystic kidney disease. Herein, we provide an overview of the potential roles of AI-assisted medical image analysis for the diagnosis and management of CKD.

Deep-learning segmentation of ultrasound images for automated calculation of the hydronephrosis area to renal parenchyma ratio

Purpose

We investigated the feasibility of measuring the hydronephrosis area to renal parenchyma (HARP) ratio from ultrasound images using a deep-learning network.

Materials and Methods

The coronal renal ultrasound images of 195 pediatric and adolescent patients who underwent pyeloplasty to repair ureteropelvic junction obstruction were retrospectively reviewed. After excluding cases without a representative longitudinal renal image, we used a dataset of 168 images for deep-learning segmentation. Ten novel networks, such as combinations of DeepLabV3+ and UNet++, were assessed for their ability to calculate hydronephrosis and kidney areas, and the ensemble method was applied for further improvement. By dividing the image set into four, cross-validation was conducted, and the segmentation performance of the deep-learning network was evaluated using sensitivity, specificity, and dice similarity coefficients by comparison with the manually traced area.

Results

All 10 networks and ensemble methods showed good visual correlation with the manually traced kidney and hydronephrosis areas. The dice similarity coefficient of the 10-model ensemble was 0.9108 on average, and the best 5-model ensemble had a dice similarity coefficient of 0.9113 on average. We included patients with severe hydronephrosis who underwent renal ultrasonography at a single institution; thus, external validation of our algorithm in a heterogeneous ultrasonography examination setup with a diverse set of instruments is recommended.

Conclusions

Deep-learning-based calculation of the HARP ratio is feasible and showed high accuracy for imaging of the severity of hydronephrosis using ultrasonography. This algorithm can help physicians make more accurate and reproducible diagnoses of hydronephrosis using ultrasonography.

Segmentation and Boundary Detection of Fetal Kidney Images in Second and Third Trimesters Using Kernel-Based Fuzzy Clustering

Organ segmentation is an important step in Ultrasound fetal images for early prediction of congenital abnormalities and to estimate delivery date. In many applications of 2D medical imaging, they face problems with speckle noise and object contours. Frequent scanning of fetal leads to clinical disturbances to the fetal growth and the quantitative interpretation of Ultrasonic images also a difficult task compared to other image modalities. In the present work a three-stage hybrid algorithm has been developed to segment the US fetal kidney images for the detection of shape and contour. At the first stage the hybrid Mean Median (Hybrid MM) filter is applied to reduce the speckle noise. Then a kernel based Fuzzy C - means clustering is used to detect the shape and contour. Finally, the texture features are obtained from the segmented images. Based on the obtained texture features, the abnormalities are detected. The Gaussian Radial basis function provides an accuracy of 80% at the second and third trimesters with weighted constant ranging from 4 to 8, compared to other global kernel functions. Similarly the proposed method has an accuracy of 86% with compared to other FCM techniques.

Automation of the kidney function prediction and classification through ultrasound-based kidney imaging using deep learning

Prediction of kidney function and chronic kidney disease (CKD) through kidney ultrasound imaging has long been considered desirable in clinical practice because of its safety, convenience, and affordability. However, this highly desirable approach is beyond the capability of human vision. We developed a deep learning approach for automatically determining the estimated glomerular filtration rate (eGFR) and CKD status. We exploited the transfer learning technique, integrating the powerful ResNet model pretrained on an ImageNet dataset in our neural network architecture, to predict kidney function based on 4,505 kidney ultrasound images labeled using eGFRs derived from serum creatinine concentrations. To further extract the information from ultrasound images, we leveraged kidney length annotations to remove the peripheral region of the kidneys and applied various data augmentation schemes to produce additional data with variations. Bootstrap aggregation was also applied to avoid overfitting and improve the model's generalization. Moreover, the kidney function features obtained by our deep neural network were used to identify the CKD status defined by an eGFR of <60 ml/min/1.73 m2. A Pearson correlation coefficient of 0.741 indicated the strong relationship between artificial intelligence (AI)- and creatinine-based GFR estimations. Overall CKD status classification accuracy of our model was 85.6% -higher than that of experienced nephrologists (60.3%-80.1%). Our model is the first fundamental step toward realizing the potential of transforming kidney ultrasound imaging into an effective, real-time, distant screening tool. AI-GFR estimation offers the possibility of noninvasive assessment of kidney function, a key goal of AI-powered functional automation in clinical practice.

Ultrasound screening for pyelectasis in pregnant women. Clinical necessity or "art for art's sake"?

Renal pelvicalyceal dilatation is caused by urine retention in the upper urinary tract. It is referred to as pyelectasis in medical literature. This term does not indicate the cause that leads to the dilatation of and urine retention in the renal pelvicalyceal system. Mild pelvicalyceal dilatation during pregnancy is usually considered to be physiological in nature – it can occur in up to 90% of pregnant women. Retention is more common in the right kidney, in primigravidae, in the second half of pregnancy and in multiple pregnancies. Pyelectasis during pregnancy rarely causes clinical symptoms and often does not require treatment. Nevertheless, urine retention in the renal pelvicalyceal system is conducive to the development of asymptomatic bacteriuria and may be a risk factor for recurrent urinary tract infections, pyelonephritis and acute kidney failure; it may also cause renal colic. In consequence, this condition can lead to intrauterine infection and premature labor in the pregnant woman and to prematurity, anemia, congenital pneumonia or sepsis in the child. In a study conducted at the 3^rd Department of Gynecology of the Medical University of Lublin it was concluded that unilateral pyelectasis of more than 20 cm³ is associated with a significant increase in the risk of asymptomatic bacteriuria. This volume corresponds to grade 3 and/or 4 pelvicalyceal dilatation according to the Society for Fetal Urology/European Federation of Societies for Ultrasound in Medicine and Biology classification. The pyelectasis volume measuring method using three-dimensional ultrasound scanning included in the criteria for the assessment of asymptomatic bacteriuria was assessed as sensitive and specific. The ultrasound-based evaluation of the kidneys for the presence of pyelectasis and its grade in pregnant women has some clinical implications. It allows for identifying cases with an increased risk of asymptomatic bacteriuria, which requires treatment in pregnant women. Screening during pregnancy for pyelectasis seems to be important in preventing asymptomatic bacteriuria from progressing to symptomatic urinary tract infection.

Sonography of tumors and tumor-like lesions that mimic carcinoma of the urinary bladder

One of the basic abdominal organs that is assessed during transabdominal ultrasound examination for urological reasons is the urinary bladder. The bladder must be filled with urine. This is a prerequisite for a reliable assessment and, at the same time, an acoustic window in examining adjacent structures and organs, for instance the prostate gland. In some cases, doubts occur with respect to the nature of lesions detected. The paper presents anatomic lesions, defects and pathologies which might be erroneously interpreted as tumors of the urinary bladder, i.e. transitional cell carcinoma of the urinary bladder. The following lesions are discussed: 1) anatomic defects (including urachus remnants, ligaments that stabilize the bladder or cyst in the opening of the ureter into the bladder – ureterocele); 2) tumor- like lesions in the lumen of the urinary bladder (such as blood clots, fungus balls, stones or foreign bodies); 3) bladder wall pathologies (i.e. cystitis or endometriosis), focal decidual transformation of stromal cells or inflammatory pseudotumor; 4) lesions impressing on the bladder from the outside (the mesentery of the sigmoid colon, the bowel, pathological lesions in organs adjacent to the urinary bladder, inflammatory infiltration, vasogenic compression of the bladder, pelvic lipomatosis, pathological lesions of the pubic symphysis); 5) postoperative lesions. All these lesions may mimic carcinoma of the urinary bladder in sonography. Bearing this fact in mind is significant in establishing a diagnosis. Due to the malignant character of carcinoma of the urinary bladder and the need for aggressive surgical treatment, a correct diagnosis of this disease is essential for patients, particularly because the lack of adequate treatment and delayed treatment considerably affect prognosis.

Errors and mistakes in ultrasound diagnostics of the thyroid gland

Ultrasound examination of the thyroid gland permits to evaluate its size, echogenicity, margins, and stroma. An abnormal ultrasound image of the thyroid, accompanied by other diagnostic investigations, facilitates therapeutic decision-making. The ultrasound image of a normal thyroid gland does not change substantially with patient's age. Nevertheless, erroneous impressions in thyroid imaging reports are sometimes encountered. These are due to diagnostic pitfalls which cannot be prevented by either the continuing development of the imaging equipment, or the growing experience and skill of the practitioners. Our article discusses the most common mistakes encountered in US diagnostics of the thyroid, the elimination of which should improve the quality of both the ultrasound examination itself and its interpretation. We have outlined errors resulting from a faulty examination technique, the similarity of the neighboring anatomical structures, and anomalies present in the proximity of the thyroid gland. We have also pointed out the reasons for inaccurate assessment of a thyroid lesion image, such as having no access to clinical data or not taking them into account, as well as faulty qualification for a fine needle aspiration biopsy. We have presented guidelines aimed at limiting the number of misdiagnoses in thyroid diseases, and provided sonograms exemplifying diagnostic mistakes.