Introduction
Methods
Patients
Patients (n = 48) | |
Male | 32 (67) |
Age (years) | 68.4 ± 11.7 |
Body weight (kg) | 61.1 ± 13.5 |
Body mass index (kg/m2) | 23.0 ± 4.1 |
Neurological diseases | 27 (56) |
Parkinson's disease | 4 |
Dementia with Lewy bodies | 2 |
Familial amyloid polyneuropathy | 6 |
Other neurological diseases including progressive supranuclear palsy | 15 (44) |
Heart diseases | 21 |
Chronic heart failure | 13 |
Arrhythmia | 5 |
Cardiomyopathy | 3 |
Reduced cardiac uptake | 17 (35) |
Left ventricular ejection fraction | 56.1 ± 17.3 |
123I-MIBG imaging
Planar image analysis
Segmentation based on CNN
Registration
Heart segmentation
CNN-based calculation of average heart counts and WR
Comparison of CNN-based and conventional quantitation
Statistical analysis
Results
Segmentation on images using CNN
SPECT H CNN versus planar H BC , planar H, and planar HMR
SPECT WR CNN versus planar WR BC , planar WR NC , and planar WR HMR
Planar WRBC | Total | ||
---|---|---|---|
< 34% | ≥ 34% | ||
(A) | |||
SPECT WRCNN | |||
≤ 30% | 23 (48.9%) | 1 (2.1%) | 24 (51.1%) |
> 30% | 5 (10.6%) | 18 (38.3%) | 23 (48.9%) |
Total | 28 (59.6%) | 19 (40.4%) | 47 |
Planar WRNC | Total | ||
---|---|---|---|
< 30.1% | ≥ 30.1% | ||
(B) | |||
SPECT WRCNN | |||
≤ 30% | 18 (38.3%) | 1 (2.1%) | 19 (40.4%) |
> 30% | 9 (19.2%) | 19 (40.4%) | 28 (59.6%) |
Total | 27 (57.5%) | 20 (42.5%) | 47 |
Planar WRHMR | Total | ||
---|---|---|---|
< 14.2% | ≥ 14.2% | ||
(C) | |||
SPECT WRCNN | |||
≤ 30% | 18 (38.3%) | 1 (2.1%) | 19 (40.4%) |
> 30% | 12 (25.5%) | 16 (34.0%) | 28 (59.6%) |
Total | 30 (63.8%) | 17 (36.2%) | 47 |