Manual Segmentation Errors in Medical Imaging. Proposing a Reliable Gold Standard

AbstractMedical imaging for forensic age estimation in living adolescents and young adults continues to be controversial and a subject of discussion. Because age estimation based on medical imaging is well studied, it is the current gold standard. However, large disparities exist between the centers conducting age estimation, both between and within countries. This review provides an overview of the most common approaches applied in Europe, with case examples illustrating the differences in imaging modalities, in staging of development, and in statistical processing of the age data. Additionally, the review looks toward the future because several European research groups have intensified studies on age estimation, exploring four strategies for optimization: (1) increasing sample sizes of the reference populations, (2) combining single-site information into multifactorial information, (3) avoiding ionizing radiation, and (4) conducting a fully automated analysis.

Download Full-text

Performance of Forced-Alignment Algorithms on Children's Speech

Journal of Speech Language and Hearing Research ◽

10.1044/2020_jslhr-20-00268 ◽

2021 ◽

pp. 1-10

Author(s):

Tristan J. Mahr ◽

Visar Berisha ◽

Kan Kawabata ◽

Julie Liss ◽

Katherine C. Hustad

Keyword(s):

Gold Standard ◽

Acoustic Measurement ◽

Manual Segmentation ◽

Speech Sample ◽

Older Children ◽

Adaptive Training ◽

Alignment Algorithms ◽

Child Speech ◽

Speech Recognition Engine ◽

Children's Speech

Purpose Acoustic measurement of speech sounds requires first segmenting the speech signal into relevant units (words, phones, etc.). Manual segmentation is cumbersome and time consuming. Forced-alignment algorithms automate this process by aligning a transcript and a speech sample. We compared the phoneme-level alignment performance of five available forced-alignment algorithms on a corpus of child speech. Our goal was to document aligner performance for child speech researchers. Method The child speech sample included 42 children between 3 and 6 years of age. The corpus was force-aligned using the Montreal Forced Aligner with and without speaker adaptive training, triphone alignment from the Kaldi speech recognition engine, the Prosodylab-Aligner, and the Penn Phonetics Lab Forced Aligner. The sample was also manually aligned to create gold-standard alignments. We evaluated alignment algorithms in terms of accuracy (whether the interval covers the midpoint of the manual alignment) and difference in phone-onset times between the automatic and manual intervals. Results The Montreal Forced Aligner with speaker adaptive training showed the highest accuracy and smallest timing differences. Vowels were consistently the most accurately aligned class of sounds across all the aligners, and alignment accuracy increased with age for fricative sounds across the aligners too. Conclusion The best-performing aligner fell just short of human-level reliability for forced alignment. Researchers can use forced alignment with child speech for certain classes of sounds (vowels, fricatives for older children), especially as part of a semi-automated workflow where alignments are later inspected for gross errors. Supplemental Material https://doi.org/10.23641/asha.14167058

Download Full-text

PV Modeling of Medical Imaging Systems

Benford's Law ◽

10.23943/princeton/9780691147611.003.0018 ◽

2015 ◽

Author(s):

John Chiverton ◽

Kevin Wells

Keyword(s):

Medical Imaging ◽

Gold Standard ◽

Ground Truth ◽

Imaging Data ◽

Nuclear Medicine Imaging ◽

Phantom Imaging ◽

Positron Emission ◽

Pet Ct ◽

Many Sources

This chapter applies a Bayesian formulation of the Partial Volume (PV) effect, based on the Benford distribution, to the statistical classification of nuclear medicine imaging data: specifically Positron Emission Tomography (PET) acquired as part of a PET-CT phantom imaging procedure. The Benford distribution is a discrete probability distribution of great interest for medical imaging, because it describes the probabilities of occurrence of single digits in many sources of data. The chapter thus describes the PET-CT imaging and post-processing process to derive a gold standard. Moreover, this chapter uses it as a ground truth for the assessment of a Benford classifier formulation. The use of this gold standard shows that the classification of both the simulated and real phantom imaging data is well described by the Benford distribution.

Download Full-text

Evaluating estimation techniques in medical imaging without a gold standard: experimental validation

10.1117/12.480330 ◽

2003 ◽

Cited By ~ 5

Author(s):

John W. Hoppin ◽

Matthew A. Kupinski ◽

Donald W. Wilson ◽

Todd E. Peterson ◽

Benjamin Gershman ◽

...

Keyword(s):

Medical Imaging ◽

Gold Standard ◽

Experimental Validation ◽

Estimation Techniques ◽

Without A Gold Standard

Download Full-text

Fully Automated GMP-Compliant Synthesis of [18F]FE-PE2I

Pharmaceuticals ◽

10.3390/ph14070601 ◽

2021 ◽

Vol 14 (7) ◽

pp. 601

Author(s):

Klas Bratteby ◽

Charlotte Lund Denholt ◽

Szabolcs Lehel ◽

Ida Nymann Petersen ◽

Jacob Madsen ◽

...

Keyword(s):

Medical Imaging ◽

Diagnostic Imaging ◽

Shelf Life ◽

Dopamine Transporter ◽

Gold Standard ◽

First Generation ◽

Imaging Techniques ◽

Radiochemical Yield ◽

Factors Affecting ◽

Relevant Factors

In the struggle to understand and accurately diagnose Parkinson′s disease, radiopharmaceuticals and medical imaging techniques have played a major role. By being able to image and quantify the dopamine transporter density, noninvasive diagnostic imaging has become the gold standard. In the shift from the first generation of SPECT tracers, the fluorine-18-labeled tracer [18F]FE-PE2I has emerged as the agent of choice for many physicians. However, implementing suitable synthesis for the production of [18F]FE-PE2I has proved more challenging than expected. Through a thorough analysis of the relevant factors affecting the final radiochemical yield, we were able to implement high-yielding fully automated GMP-compliant synthesis of [18F]FE-PE2I on a Synthera®+ platform. By reaching RCYs up to 62%, it allowed us to isolate 25 GBq of the formulated product, and an optimized formulation resulted in the shelf life of 6 h, satisfying the increased demand for this radiopharmaceutical.

Download Full-text

Is Manual Segmentation the Real Gold Standard for Tooth Segmentation? A Preliminary in vivo Study Using Cone-beam Computed Tomography Images

Meandros Medical And Dental Journal ◽

10.4274/meandros.galenos.2021.14632 ◽

2021 ◽

Vol 22 (3) ◽

pp. 263-273

Author(s):

Sercan Sabancı ◽

Elif Şener ◽

Rukiye Irmak Turhal ◽

Barış Oğuz Gürses ◽

Figen Gövsa ◽

...

Keyword(s):

Computed Tomography ◽

Cone Beam Computed Tomography ◽

Gold Standard ◽

Cone Beam ◽

Manual Segmentation ◽

The Real ◽

Computed Tomography Images ◽

Tooth Segmentation

Download Full-text

Estimation in Medical Imaging without a Gold Standard

Academic Radiology ◽

10.1016/s1076-6332(03)80372-0 ◽

2002 ◽

Vol 9 (3) ◽

pp. 290-297 ◽

Cited By ~ 25

Author(s):

Matthew A. Kupinski ◽

John W. Hoppin ◽

Eric Clarkson ◽

Harrison H. Barrett ◽

George A. Kastis

Keyword(s):

Medical Imaging ◽

Gold Standard ◽

Without A Gold Standard

Download Full-text

A new method for quantitative assessment of hand muscle volume and fat in magnetic resonance images

BMC Rheumatology ◽

10.1186/s41927-020-00170-3 ◽

2020 ◽

Vol 4 (1) ◽

Author(s):

Andreas Friedberger ◽

Camille Figueiredo ◽

Tobias Bäuerle ◽

Georg Schett ◽

Klaus Engelke

Keyword(s):

Gold Standard ◽

Quantitative Assessment ◽

Fat Content ◽

Magnetic Resonance Images ◽

Muscle Volume ◽

Manual Segmentation ◽

Automated Method ◽

Average Analysis ◽

Hand Muscle ◽

Muscle Segmentation

Abstract Background Rheumatoid arthritis (RA) is characterized by systemic inflammation and bone and muscle loss. Recent research showed that obesity facilitates inflammation, but it is unknown if obesity also increases the risk or severity of RA. Further research requires an accurate quantification of muscle volume and fat content. Methods The aim was to develop a reproducible (semi) automated method for hand muscle segmentation and quantification of hand muscle fat content and to reduce the time consuming efforts of manual segmentation. T1 weighted scans were used for muscle segmentation based on a random forest classifier. Optimal segmentation parameters were determined by cross validation with 30 manually segmented hand datasets (gold standard). An operator reviewed the automatically created segmentation and applied corrections if necessary. For fat quantification, the segmentation masks were automatically transferred to MRI Dixon sequences by rigid registration. In total 76 datasets from RA patients were analyzed. Accuracy was validated against the manual gold standard segmentations. Results Average analysis time per dataset was 10 min, more than 10 times faster compared to manual outlining. All 76 datasets could be analyzed and were accurate as judged by a clinical expert. 69 datasets needed minor manual segmentation corrections. Segmentation accuracy compared to the gold standard (Dice ratio 0.98 ± 0.04, average surface distance 0.04 ± 0.10 mm) and reanalysis precision were excellent. Intra- and inter-operator precision errors were below 0.3% (muscle) and 0.7% (fat). Average Hausdorff distances were higher (1.09 mm), but high values originated from a shift of the analysis VOI by one voxel in scan direction. Conclusions We presented a novel semi-automated method for quantitative assessment of hand muscles with excellent accuracy and operator precision, which highly reduced a traditional manual segmentation effort. This method may greatly facilitate further MRI image based muscle research of the hands.

Download Full-text