moRphology - dEep Learning Imaging Cells (RELIC) - to Differentiate Between Normal and Pathological Kidney Exfoliated Cells

Abstract

Chronic kidney disease (CKD) is characterised by progressive loss of kidney function leading to kidney failure. Significant kidney damage can occur before symptoms are detected. Currently, kidney tissue biopsy is the gold standard for evaluation of renal damage and CKD severity. This study explores how to precisely quantify morphology characteristics of kidney cells exfoliated into urine, with a view to establish a future urine-based non-invasive diagnostic for CKD. We report the development of a novel deep learning method, which was able to discover a RELIC (moRphology dEep Learning Imaging Cells) signature that can differentiate between kidney cells exfoliated in human urine of and CKD patients with varying degree of kidney damage and non-CKD controls. Exfoliated proximal tubule cells (PTCs) originating from kidneys were isolated from the urine of patients with different levels of kidney damage using previously published methods. An advanced combination of artificial intelligence techniques, deep learning, swarm intelligence, and discriminative analysis was used to discover a RELIC signature in brightfield microscopy images of exfoliated PTCs. Kidney damage in the study subjects was characterised by assessing kidney tissues obtained through a nephrectomy or kidney biopsy. A special deep learning algorithm was developed and trained to create a predictive tool. The algorithm was then used to analyse data from patients with normal and fibrotic kidneys. Data were then classified according to different groups (healthy or fibrosis) and clustering of the training and validation cells was determined for model validation. We developed a novel deep learning method, to obtain RELIC signatures and identify specific deep morphological features which can be used to differentiate urinary PTC cells shed by people with CKD (confirmed by tissue histology obtained from an invasive kidney biopsy) from those without CKD, with a discriminatory accuracy of 82%. We identified a RELIC signature which can be used on a collection of bright field images of exfoliated urinary PTCs to create a predictive tool and differentiate between normal and pathological kidney cells. This study, for the first time, provides a proof of concept that urinary exfoliated tubule cells in patients with kidney fibrosis and healthy controls differ in appearance (morphology) as observed under a basic brightfield microscope. The results suggest that morphological signatures of exfoliated PTCs have the potential to serve as a non-invasive marker of kidney fibrosis.

Automated identification of aneuploid cells within the inner cell mass of an embryo using a numerical extraction of morphological signatures

“…To identify features, the CNN uses convolutional layers successively applied to an input image while in traditional algorithms features were obtained by application of mathematical definition of features (Campbell, et al, 2019). A sequence of three nets (Net1, Net2 and Net3) were designed for this work(Habibalahi, Bertoldo, Mahbub, Campbell, Goss, Ledger, Gilchrist, Wu and Goldys, 2021, Habibalahi, Campbell, Mahbub, Anwer, Nguyen, Gill, Wong, Pollock, Saad and Goldys, 2022). Each CNN generated a separate feature collection and subsequently all features were collated.…”

“…Each CNN generated a separate feature collection and subsequently all features were collated. Each net has specific configurations and resolutions, which leads to detailed and comprehensive image feature extraction(Habibalahi, Bertoldo, Mahbub, Campbell, Goss, Ledger, Gilchrist, Wu and Goldys, 2021, Habibalahi, Campbell, Mahbub, Anwer, Nguyen, Gill, Wong, Pollock, Saad and Goldys, 2022). Net 1 has a deep depth including 153 convolutional layers whose filters were taken from ResNet (He, et al, 2016).…”

“…Briefly, the simple artificial agents (here feature subsets) are moving satisfying a pre-defined development rule by maximizing Fisher Distance (FD) as a measure of optimization. Initially, an agent population (here feature subsets) are created, and corresponding FD is calculated for each agent(Habibalahi, Campbell, Mahbub, Anwer, Nguyen, Gill, Wong, Pollock, Saad and Goldys, 2022). Next, agents are continually revised based on the pre-defined strategy of evolution till satisfying the convergence of FD.…”

“…Image augmentation also expands the data set which improves the deep learning performance by enhancing training images to discover stronger DMS. Embryo images and their corresponding reflections, in this study, were rotated at various angles (45°, 90°, 135°, 180°) (Habibalahi, Bertoldo, Mahbub, Campbell, Goss, Ledger, Gilchrist, Wu and Goldys, 2021, Habibalahi, Campbell, Mahbub, Anwer, Nguyen, Gill, Wong, Pollock, Saad and Goldys, 2022).…”

Unique Deep Radiomic Signature Shows NMN Treatment Reverses Morphology of Oocytes from Aged Mice

“…Recently, artificial intelligence (AI) has been widely applied to objective biomedical image assessment for disease diagnosis and monitoring to enable the precise customization of treatment plans [ 5 , 6 , 7 , 8 , 9 ]. Deep learning strategies (machine learning algorithms that use multiple layers to progressively extract higher-level features from data) have been used to interpret electroencephalogram (EEG), electrocardiogram (ECG), magnetoencephalography (MEG), and magnetic resonance imaging (MRI) data, to improve reliability and precision [ 10 ].…”