UriSed Technology – A Standardised Automatic Method of Urine Sediment Analysis
Table 1: Diagnostic Performance Characteristics of UriSed TP
TN
Particle RBC WBC NEC EPI
Yeast CaOx PAT HYA BAC
Particle RBC WBC NEC EPI
Yeast CaOx PAT HYA BAC
Study I 495 497 82
154 65 27 61 43
325
% PPV Study I 87 89 53 74 37 18 33 20 64
Study II 539 404 126 358 -
30
445 558 -
Study II 79 68 19 69 -
14 57 68 -
Study I 224 272 667 675 713 753 658 724 293
% NPV Study I 65 78 91 96 97 99 91 98 73
Study II 467 601 570 648 -
1010 304 277 -
Study II 85 95 97 91 -
99 68 68 -
NPV = negative predictive value; PPV = positive predictive value; TN = true negative; TP = true positive.
would be very hard to explain how one can distinguish between a television and a computer monitor.
The AIEM is trained to recognise particles in the raw image itself without extracting any information from the image as feature parameters. The main advantage of this technology is that it uses a very wide range of information to classify different particles in the image. The image naturally contains information about the size, shape, contrast, texture and many other characteristics of the particles, too. By scanning the original whole field of view image itself instead of working from extracted information, the AIEM receives more input information than is the case with the simplified, feature parameter-based methods of other technologies, thus providing a more accurate recognition algorithm.
Although the efficiency and diagnostic performance of the instrument have already been proven by different studies, there is still significant potential in the improvement of the AIEM. There have been five versions of the AIEM developed in the last three years and the performance of each module has been better than the previous. The manufacturer continuously develops the performance of the AIEM software; new AIEMs can be implemented at any time by a routine software upgrade on the already installed units as well.
UriSed results provide a direct link to the clinical meaning by producing whole field of view microscopic images in which no information can be lost. Two images of different urine samples taken by UriSed can be seen on Figures 4 and 5. The AIEM is able to automatically classify and count numerous urine sediment particles in the images, such as red blood cells (RBC), white blood cells (WBC), squamous epithelial cells (EPI), non-squamous epithelial cells (NEC), hyaline casts (HYA), pathological casts (PAT), calcium-oxalate crystals (CaOx), uric acid crystals (URI), triple-phosphate crystals (TRI), bacteria (BAC), yeast (YEA), sperm (SPRM) and mucus (MUC). The morphology of sediment particles is shown in the whole field of view images. For
EUROPEAN INFECTIOUS DISEASE Figure 3: Auto Image Evaluation Module
% Sensitivity Study I 81 87 52 86 74 82 47 71 75
Linear Weighted κ Study I 0.65 0.72 0.41 0.67 0.46 0.24 0.26 0.29 0.37
Study II 87 92 89 85 -
79 76 81 -
Study II 0.68 0.69 0.16 0.62 -
0.21 0.22 0.33 -
% Specificity Study I 76 81 91 93 87 86 84 85 61
Odds Ratio Study I 12.9 27.6 10.7 73.3 18.5 27.4 4.9
14.0 4.7
Study II 76 76 52 80 -
85 41 51 -
Study II 21.2 36.4 8.8
22.7 -
21.3 2.8 4.4 -
Input: raw whole viewed image
Auto Image Evaluation Module
Output: evaluated whole viewfield image
Figure 4: UriSed Image – Sample with Squamous Epithelial Cells, Calcium Oxalate Dihydrate Crystals, Red Blood Cells and White Blood Cells
EPI
EPI CaOxd EPI EPI CaOxd
CaOxd
EPI CaOxd EPI WBC EPI CaOxd EPI
RBC
EPI WBC CaOxd EPI EPI CaOxd CaOxd CaOxd EPI
CaOxd = calcium oxalate dihydrate; EPI = squamos epithelial cells; WBC = white blood cells.
example, acanthocytes, other isomorphic erythrocytes and different types of pathological casts may be identified, which provides extra information for doctors. In addition, not only the morphology of individual urine particles, but also the general look of urine sediment can be investigated in these images, which gives a broader spectrum
141
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92