Pediatric thoracic dynamic MRI, six organs

Dynamic MR of one healthy child's thorax and upper abdomen, at end expiration or end inspiration, with the lungs, liver, spleen and kidneys segmented.

28 test casesthorax, abdomenimage MR

Look at the images

Read before comparing (2)

Cases. Both images of each PediatricThoraxMR pair, as delivered, on that pair's side of the split, so a child's two phases are in one split.
Images. Free-breathing 3 T true-FISP, 40 sagittal slices at each of about 35 positions over several breathing cycles. The authors' OFx method picks one slice per position and phase to build one cycle, then standardises each volume's intensities. A volume stacks slices from different cycles, so steps between neighbouring slices belong to the images. The archive holds only the end-expiration and end-inspiration volumes.

		Accuracy			Cost
#	Method	Dice	Surface Dice	HD95 mm	Time s	RAM GB	GPU GB	Images
1	KonfAI segmentation app · TotalSegmentator (total_mr), the authors' packageGPUapp	0.750 ±0.019	0.445 ±0.017	14.34 ±1.15	23	11.1	3.0
2	KonfAI segmentation app · TotalSegmentator (total_mr), KonfAI appGPUapp	0.737 ±0.020	0.422 ±0.017	40.58 ±10.98	27	2.5	4.0
3	KonfAI segmentation app · MRSegmentator, KonfAI appGPUapp	0.067 ±0.011	0.024 ±0.003	108.41 ±7.08		3.1	5.6

The order averages each method's ranks on Dice, Surface Dice, HD95 mm, so the first one is not always the best in every column. ± is the 95% interval of the mean over the 28 cases. Two neighbours are tied when a paired Wilcoxon test on the shared cases does not find the upper one better on Dice (5%): the places a method could hold are then shown under its own. The best value of each column is in colour. Click a row to see its images, a column title to sort by it. Time and memory measured on Workstation.

1 result not ranked: an older run, cases still missing, or a failure on every case

		Accuracy			Cost
#	Method	Dice	Surface Dice	HD95 mm	Time s	RAM GB	GPU GB	Machine	Images
	KonfAI segmentation app · TotalSegmentator (total_mr), KonfAI appGPUappolder run	0.737 ±0.020	0.422 ±0.017	39.88 ±10.63	6	2.2	7.3	Laptop

Dice: how much the organ contours overlap, 1 is perfect. Surface Dice: the share of the contours' surfaces within a tolerance of each other, 1 is perfect. HD95: distance between the contours' surfaces.

Case by case

Measure

Dice	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26	27	28
KonfAI segmentation app · TotalSegmentator (total_mr), the authors' package	0.774	0.800	0.832	0.857	0.708	0.743	0.747	0.806	0.754	0.772	0.776	0.709	0.652	0.803	0.729	0.754	0.638	0.675	0.750	0.773	0.735	0.759	0.698	0.719	0.776	0.743	0.752	0.778
KonfAI segmentation app · TotalSegmentator (total_mr), KonfAI app	0.763	0.783	0.812	0.838	0.765	0.768	0.699	0.770	0.731	0.738	0.749	0.679	0.619	0.718	0.754	0.765	0.603	0.641	0.710	0.723	0.743	0.765	0.759	0.728	0.775	0.761	0.728	0.747
KonfAI segmentation app · MRSegmentator, KonfAI app	0.056	0.083	0.080	0.082	0.063	0.057	0.064	0.069	0.085	0.114	0.066	0.073	0.057	0.056	0.017	0.035	0.065	0.075	0.085	0.160	0.071	0.058	0.025	0.045	0.033	0.055	0.076	0.076

Click a score to look at that result: the image, the truth in green and the method's map in red. Underlined scores have their image; for the others you see the case without the result.

About the data

The first test caseClick to look at every case.

Truth: contours by a program, verified by people the dataset's masks at both phases: left lung (1), right lung (2), liver (3), spleen (4), left kidney (5), right kidney (6)
Source: doi.org/10.5061/dryad.vmcvdnczf
www.mipg.upenn.edu/Vnews/mipg_data_share/Thorax_dMRI.html
doi.org/10.1117/12.3006807
doi.org/10.1016/j.media.2021.102088
Licence: CC0 1.0

24 cases for tuning, never tested

VGC264/fixed, VGC264/moving, VGC265/fixed, VGC265/moving, VGC266/fixed, VGC266/moving, VGC484/fixed, VGC484/moving, VGC478/fixed, VGC478/moving, VGC442/fixed, VGC442/moving, VGC338/fixed, VGC338/moving, VGC272/fixed, VGC272/moving, VGC397/fixed, VGC397/moving, VGC322/fixed, VGC322/moving, VGC285/fixed, VGC285/moving, VGC430/fixed, VGC430/moving