Towards Clinical Integration of Deep Learning-Based Classification of Urinary Sediment Particles from Digital Microscopy Images: A Prospective Study.

The efficient workflow to decrease the manual microscopic examination of urine sediment using on-screen review of images

The microscopic examination of urine sediment is a common diagnostic tool taught to medical students, medical technologists, and others. The urine microscopic exam is difficult to teach because supervised instruction and textbook-based teaching suffer from numerous drawbacks. Here, we describe Urinalysis-Tutor, a computer program that uses digitized microscope images and computer-based teaching techniques to systematically teach the urine microscopic exam. In addition, we report the results of a 2-year study that evaluated the effectiveness of the program in 314 second year medical students who were required to use the program. The program contained two, 20-question exams. In the first year of the study (1996), one of the exams was chosen as the pretest and the other as the posttest; the pretest had to be completed before the students viewed the contents of the program, and the posttest was taken after finishing the tutorial. In 1997, the order of the two exams was reversed. In 1996, 159 students completed the study. The mean pretest score was 34% (SD, 14%), the mean posttest score was 71% (SD, 13%), and the improvement was significant (P <0.001, paired t-test). In 1997, 155 students participated. The mean pretest score was 41% (SD, 11%), the mean posttest score was 71% (SD, 13%), and the improvement was significant (P <0.001, paired t-test). The study shows that Urinalysis-Tutor helps medical students learn to interpret the microscopic appearance of urine sediment and that it is feasible to implement this tutorial in a medical school class.

Automated urine analysis poses the advantage of rapid turn around time and is suitable for initial screening of large number of samples.However, standardized manual microscopic examination of urine sediment remains valuable for a variety of differential diagnosis，such as screening the potential causes of hematuria, to predict proliferative and non-proliferative renal pathological injury. Urinary sediment scoring system is often used to differentiate acute tubular necrosis (ATN) from pre-renal acute kidney injury (AKI).The urinary podocytes serves as a marker of glomerular injury to determine the location of and to monitor activity of glomerular lesions, as well as to differentiate focal segmental glomerulosclerosis (FSGS) and minimal change disease (MCD).These useful analyses shows the importance of urine microscopy in the clinical practice of nephrology.(Chin J Lab Med, 2012，35：780-783) Key words: Kidney diseases; Kidney failure; acute; Podocytes; Urinalysis; Urine sediment; Microscopy

Several recent reports suggest that amyloidosis can be diagnosed by ultrastructural identification of fibrillar proteins in the urinary sediment. However, the authors routinely identified similar fibrils in the urinary sediments of all patients with a wide variety of renal diseases who were producing urinary casts, as well as in those of stressed healthy individuals. They therefore conclude that electron microscopic examination of the urinary sediment alone has little value in the diagnosis of amyloidosis when unaccompanied by chemical or immunologic tests.

What’s Your Diagnosis?

a) ObjectivesUrinalysis is one of the most common laboratory screening tests to detect problems in the renal and urinary system; however, they cannot detect atypical cells (Atyp.Cs). The Sysmex UF-5000, a fully automated urine particle analyzer, can detect Atyp.Cs via its Atyp.C parameter. This study aimed to compare the clinical value of the Atyp.C parameter with that of urine sediment microscopy. b) MethodA total of 471 leftover urine samples were submitted to the Department of Clinical Laboratory at the University of Tokyo Hospital for urinalysis by manual sediment microscopy examination and UF-5000 Atyp.C analysis. c) ResultOf 471 submitted samples, 117 were positive for Atyp.Cs by urine sediment and 354 samples were negative. The histological subtypes of the Atyp.Cs included 105 cases of suspected urothelial carcinoma cells, 10 suspected squamous carcinoma cells, and 2 of suspected adenocarcinoma cells. The Atyp.C values for the Atyp.C-positive and -negative groups were 2.64 ± 0.69 and 0.38 ± 0.16, respectively. The optimal Atyp.C cutoff value determined by the receiver operating characteristic curve analysis was 0.4/μL. The area under the curve was 0.856, with a sensitivity of 79.5% and specificity of 85.1%. Atyp.C values of the UF-5000 showed high predictive performance for Atyp.C-positive specimens identified by urine sediment microscopy. d) ConclusionsThis study shows that a combination of UF-5000 analysis and microscopic examination of urine sediment improves Atyp.C detection in urine sediment analysis. These results suggest that Atyp.C measured by UF-5000 could be a useful screening parameter in routine testing of urine samples.

A Comparison of Urinary Albumin–Total Protein Ratio to Phase-Contrast Microscopic Examination of Urine Sediment for Differentiating Glomerular and Nonglomerular Bleeding

A review on various methods for recognition of urine particles using digital microscopic images of urine sediments

SUMMARY A retrospective survey of complete urinalyses performed on dogs and cats hospitalized at the University of Minnesota Veterinary Teaching Hospital was performed to determine the frequency of abnormal findings in urine sediment, compared with normal and abnormal macroscopic findings. One thousand consecutive urine samples collected from dogs and 1,000 consecutive urine samples collected from cats were studied. Results indicated that failure to examine urine sediment of macroscopically normal samples would have yielded false-negative results in 16.5% of the canine patients and 5.7% of the feline patients. Microscopic findings in macroscopically normal canine samples consisted primarily of pyuria and bacteriuria, whereas in cats they consisted of hematuria and bacteriuria. Microscopic abnormalities were detected in almost 50% of the canine and feline samples with macroscopic abnormalities. The results of this evaluation confirm the value of routine microscopic examination of urine sediment as a component of complete urinalysis.

Microscopic examination of urine sediment is a basic, common method of detecting diseases of the urinary tract. We experienced a case involving a patient who had developed a fever after undergoing a urinary-diversion operation. Results of the microscopic examination of the urine collected from the urostomy pouch of the patient showed a food debris-like component. Based on this finding, we suspected a fistula between the urinary and intestinal tracts. However, after performing an experiment to verify the results, we determined that no fistula was present. Instead, we discovered that the food debris-like component originated from the urostomy skin barrier. To our knowledge, ours is the first report in the literature to demonstrate that the urostomy skin barrier can dissolve and mimic food debris in urine sediment, leading to incorrect assumptions regarding the presence of fistulas. We believe that it is important to derive a correct diagnosis when an unfamiliar component is observed in urinary sediment. We further believe that the findings from this case are valuable to professionals who administer clinical treatment and perform laboratory testing.

Bacterial infection of the urinary tract is a common disorder in dogs and cats. Although microscopic examination of urine sediment is routinely used to screen for infection, this test can lack sensitivity or require expertise. A reliable in-clinic screening test would be a useful adjunct for the identification of dogs and cats with bacterial urinary tract infection (UTI). That a catalase-based urine test (Accutest Uriscreen™) is a more sensitive screening test for UTI in dogs and cats than urine microscopic sediment examination. One hundred and sixty client-owned dogs and cats. Surplus urine from animals presented to a veterinary teaching hospital was used in this prospective observational study. A routine urinalysis, aerobic bacterial culture, and the Uriscreen test were performed on cystocentesis samples. Sensitivity and specificity with 95% confidence intervals and positive and negative likelihood ratios were calculated for Uriscreen and microscopic sediment examination using culture results as the gold standard. Bacterial culture was positive in 27/165 (16.4%) samples. The sensitivity, specificity, and positive and negative likelihood ratios for the Uriscreen were 89%, 71%, 3.0, and 0.15, respectively. Sensitivity, specificity, and positive and negative likelihood ratios for urine sediment microscopic examination were 78%, 90%, 7.8, and 0.24, respectively. The Uriscreen is a more sensitive screening test for UTI in dogs and cats than sediment examination; however, the urine sediment examination was more specific. A negative Uriscreen result helps exclude UTI; however, urine bacterial culture is still necessary to exclude or confirm UTI in all cases.

Urinary sediment is an important part of routine urine test, which plays an irreplaceable role in the diagnosis of diseases, monitoring of treatment effect, and prognosis judgment [1]. Through the results of urine dry chemistry and microscopic examination of urinary sediment, we inter-preted and analyzed the clinical significance of urinary casts in urinary sediment. In patients with new urinary system diseases abnormal urine results appear earlier than changes in serum renal function indicators, especially when the urine sediment shows typical casts, which can provide an important basis for clinical diagnosis. Clinical laboratory personnel should attach great importance to the morphological examination of urinary sediment and master the diagnostic significance of the formed components of urinary sediment for various diseases, so as to better assist clinical disease diagnosis.

Objective:Microscopic examination of urine sediment is necessary for evaluation of renal and urinary tract diseases. In this study, we evaluated and compared analytic and diagnostic performances of DIRUI FUS-200 and a new image-based automated urine sediment analyzer Urised 3.Method:A total of 440 urine samples, submitted to our laboratory, were evaluated by two automated urine sediment analyzers and a standardized manual microscopy. Precision, linearity and method comparison studies were performed according to CLSI guidelines.Results:Considering the red blood cell (RBC) and white blood cell (WBC) counts, strong correlations existed between FUS-200 and manual microscopy (r=0.993 vs 0.861), Urised 3 and manual microscopy (r=0.962 vs 0.818), FUS200 and Urised 3 (r=0.961 vs 0.961). Clinical non-concordance ranged from 7% to 14.16% among all methods.Conclusions:The concordance between the analyzers and manual microscopy for WBC was better than that of RBC. The concordance between the two analyzers was better for WBC and RBC, with respect to the manual microscopy. Although the Urised 3, FUS-200 and manual microscopy counts were in agreement; confirmation of the results of automated analyzers with manual microscopy is particularly helpful, for pathological samples with near cut-off values.

To the Editor.— I read the letter from Robert M. Donauer, MD (240:2044, 1978), with much interest. There are two questions I would like to ask the author: Were the urine samples centrifuged before the dipstick tests? What kinds of findings were there among those 12% showing negative chemical screening tests and positive microscopic findings? My own study of roughly 1,000 random urine samples from an active community hospital showed that there is an approximate 6.5% chance that the dipstick will miss the socalled significant microscopic finding. The positive microscopic findings were defined as urine sediment with more than five WBCs per high-power field or more than three RBCs per high-power field or any casts. If I apply the author's criteria to the sample population from this institution, the false-negative rate becomes 3.6%. Dr Donauer may have obtained higher percentages of negative results by dipstick if he used

Microscopic examination of urine sediment using shih-yung system requires 10ml urine volume. Pediatric patients and kidney disease patients with oliguria difficult to get 10mL urine samples, so it needs to be studied to determine the accuracy of results with smaller sample volumes. Objectives: To analyze agreement of urine sediment examination using Shih-yung system in 10 mL and 5 mL urine volume. Methods: This study was a cross-sectional analysis of 37 urine specimens that met inclusion and exclusion criteria at the central laboratory of Dr. M. Djamil Hospital from July to September 2020. Examination of erythrocyte sediment, leukocyte sediment, and non-hyaline cast using Shih-yung system was done in 10 mL and 5 mL urine volume. Data were analyzed using kappa reliability/agreement test and significant if p&lt;0.05. Results: Agreement test of erythrocyte sediment examination using Shih-yung system in 10ml and 5mL urine volume were moderate significantly (κ=0.600; p&lt;0.05), agreement of leukocyte sediment examination were moderate significantly (κ=0.571; p&lt;0.05), and agreement of nonhialin cast examination were substantial significantly (κ=0.767; p&lt;0.05). Conclusion: Agreement test of erythrocyte and leukocyte sediment examination in 10ml and 5mL urine volume is moderate, so 5ml urine volume was not recommended. Manual examination with microscope and difference in urine sample volume causing difference sediment concentration and amount of sediment per microscope field of view.Keywords: Shih-yung,  urine sediment,  urine volume