Use of a Smartphone Camera at the Bedside to Assess Adequacy of Kidney Biopsies

Smartphone-based machine learning model for real-time assessment of medical kidney biopsy

Background

Kidney biopsy is the gold-standard for diagnosing medical renal diseases, but the accuracy of the diagnosis greatly depends on the quality of the biopsy specimen, particularly the amount of renal cortex obtained. Inadequate biopsies, characterized by insufficient cortex or predominant medulla, can lead to inconclusive or incorrect diagnoses, and repeat biopsy. Unfortunately, there has been a concerning increase in the rate of inadequate kidney biopsies, and not all medical centers have access to trained professionals who can assess biopsy adequacy in real time. In response to this challenge, we aimed to develop a machine learning model capable of assessing the percentage cortex of each biopsy pass using smartphone images of the kidney biopsy tissue at the time of biopsy.

Methods

747 kidney biopsy cores and corresponding smartphone macro images were collected from five unused deceased donor kidneys. Each core was imaged, formalin-fixed, sectioned, and stained with Periodic acid-Schiff (PAS) to determine cortex percentage. The fresh unfixed core images were captured using the macro camera on an iPhone 13 Pro. Two experienced renal pathologists independently reviewed the PAS-stained sections to determine the cortex percentage. For the purpose of this study, the biopsies with less than 30% cortex were labeled as inadequate, while those with 30% or more cortex were classified as adequate. The dataset was divided into training (n=643), validation (n=30), and test (n=74) sets. Preprocessing steps involved converting High-Efficiency Image Container iPhone format images to JPEG, normalization, and renal tissue segmentation using a U-Net deep learning model. Subsequently, a classification deep learning model was trained on the renal tissue region of interest and corresponding class label.

Results

The deep learning model achieved an accuracy of 85% on the training data. On the independent test dataset, the model exhibited an accuracy of 81%. For inadequate samples in the test dataset, the model showed a sensitivity of 71%, suggesting its capability to identify cases with inadequate cortical representation. The area under the receiver-operating curve (AUC-ROC) on the test dataset was 0.80.

Conclusion

We successfully developed and tested a machine learning model for classifying smartphone images of kidney biopsies as either adequate or inadequate, based on the amount of cortex determined by expert renal pathologists. The model's promising results suggest its potential as a smartphone application to assist real-time assessment of kidney biopsy tissue, particularly in settings with limited access to trained personnel. Further refinements and validations are warranted to optimize the model's performance.

Bedside Method to Check the Adequacy of Kidney Biopsy Sample with a Smartphone Camera and Macro Lenses: A Prospective Cohort Study

Background

The utilization of smartphone-assisted evaluation is emerging in the field of histopathology. This technique improves the adequacy of samples at the bedside, avoids procedure-related complications, reduces unnecessary repeat biopsies, and saves the cost of the procedure. This study aims to compare the number of glomeruli in a renal biopsy specimen obtained by an ultrasound-guided percutaneous needle biopsy, counted at the bedside using a smartphone fitted with a 16-megapixel macro lens (Bedside method) with that observed under a light microscope after the processing of the biopsy specimen (LM method).

Materials and Methods

In this prospective cohort study, 24 consecutive adult patients (48 kidney biopsy samples) who underwent kidney biopsies were enrolled. All specimens were extracted by an ultrasound-guided percutaneous renal biopsy from the lower pole of the left kidney. Patients' demographics and clinical data were prospectively collected. The number of glomeruli in all the biopsy specimens was counted using a smartphone fitted with a 16-megapixel macro lens at the bedside (Bedside method) and subsequently under a light microscope by a pathologist after processing the biopsy specimen (LM method). Seven or more glomeruli in the specimen were considered adequate in our study.

Results

The mean age of patients at biopsy was 46.9 ± 16 years with slightly male predominance (54.2%). A total of 47 specimens were obtained from 24 patients. Of the 24 patients, 22 had native kidney biopsy and 2 had renal allograft biopsy. The average number of cores obtained per patient was 1.96. The length of core specimens ranged from 1.5 to 2 cm. A good agreement was found between bedside adequacy and slide adequacy, κ =0.684, P = 0.000. The positive agreement rate and negative agreement rate were 91.4% and 23.1%, respectively.

Conclusion

In the modern era of technology, the smartphone is a good tool to evaluate the adequacy of biopsy specimens at the bedside.

A web-based tool for real-time adequacy assessment of kidney biopsies

The escalating incidence of kidney biopsies providing insufficient tissue for diagnosis poses a dual challenge, straining the healthcare system and jeopardizing patients who may require rebiopsy or face the prospect of an inaccurate diagnosis due to an unsampled disease. Here, we introduce a web-based tool that can provide real-time, quantitative assessment of kidney biopsy adequacy directly from photographs taken with a smartphone camera. The software tool was developed using a deep learning-driven automated segmentation technique, trained on a dataset comprising nephropathologist-confirmed annotations of the kidney cortex on digital biopsy images. Our framework demonstrated favorable performance in segmenting the cortex via 5-fold cross-validation (Dice coefficient: 0.788±0.130) (n=100). Offering a bedside tool for kidney biopsy adequacy assessment has the potential to provide real-time guidance to the physicians performing medical kidney biopsies, reducing the necessity for re-biopsies. Our tool can be accessed through our web-based platform: http://www.biopsyadequacy.org.

Fluorescence Imaging Using Enzyme-Activatable Probes for Detecting Diabetic Kidney Disease and Glomerular Diseases

A clear identification of the etiology of glomerular disease is essential in patients with diabetes. Renal biopsy is the gold standard for assessing the underlying nephrotic pathology; however, it has the risk for potential complications. Here, we aimed to investigate the feasibility of urinary fluorescence imaging using an enzyme-activatable probe for differentiating diabetic kidney disease and the other glomerular diseases. Hydroxymethyl rhodamine green (HMRG)-based fluorescent probes targeting gamma-glutamyl transpeptidase (GGT) and dipeptidyl-peptidase (DPP) were used. Urinary fluorescence was compared between groups which were classified by their histopathological diagnoses (diabetic kidney disease, glomerulonephritis, and nephrosclerosis) as obtained by ultrasound-guided renal biopsy. Urinary fluorescence was significantly stronger in patients with diabetic kidney disease compared to those with glomerulonephritis/nephrosclerosis after DPP-HMRG, whereas it was stronger in patients with nephrosclerosis than in patients with glomerulonephritis after GGT-HMRG. Subgroup analyses of the fluorescence performed for patients with diabetes showed consistent results. Urinary fluorescence imaging using enzyme-activatable fluorescence probes thus represents a potential noninvasive assessment technique for kidney diseases in patients with diabetes.

Safety of Obtaining an Extra Biobank Kidney Biopsy Core

Background and objectives: Kidney biopsy (KB) is the “gold standard” for the diagnosis of nephropathies and it is a diagnostic tool that presents a low rate of complications. Nowadays, biobank collections of renal tissue of patients with proven renal pathology are essential for research in nephrology. To provide enough tissue for the biobank collection, it is usually needed to obtain an extra kidney core at the time of kidney biopsy. The objective of our study is to evaluate the complications after KB and to analyze whether obtaining an extra core increases the risk of complications. Material and methods: Prospective observational study of KBs performed at Vall d’Hebron Hospital between 2019 and 2020. All patients who accepted to participate to our research biobank of native kidney biopsies were included to the study. Clinical and laboratory data were reviewed and we studied risk factors associated with complications. Results: A total of 221 patients were included, mean age 56.6 (±16.8) years, 130 (58.8%) were men, creatinine was 2.24 (±1.94) mg/dL, proteinuria 1.56 (0.506–3.590) g/24 h, hemoglobin 12.03 (±2.3) g/dL, INR 0.99 (±0.1), and prothrombin time (PT) 11.86 (±1.2) s. A total of 38 patients (17.2%) presented complications associated with the procedure: 13.1% were minor complications, 11.3% (n = 25) required blood transfusion, 1.4% (n = 3) had severe hematomas, 2.3% (n = 5) required embolization, and 0.5% (n = 1) presented arterio-venous fistula. An increased risk for complication was independently associated with obtaining a single kidney core (vs. 2 and 3 cores) (p = 0.021). Conclusions: KB is an invasive and safe procedure with a low percentage of complications. Obtaining an extra kidney core for research does not increase the risk of complications during the intervention, which remains low in concordance with previously published reports.