Detecting Multiple Myeloma via Generalized Multiple-Instance Learning

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11110%2F18%3A10376910" target="_blank" >RIV/00216208:11110/18:10376910 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/68407700:21230/18:00321266

Výsledek na webu

<a href="https://doi.org/10.1117/12.2293112" target="_blank" >https://doi.org/10.1117/12.2293112</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1117/12.2293112" target="_blank" >10.1117/12.2293112</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Detecting Multiple Myeloma via Generalized Multiple-Instance Learning

Popis výsledku v původním jazyce

We address the task of automatic detection of lesions caused by multiple myeloma (MM) in femurs or other long bones from CT data. Such detection is already an important part of the multiple myeloma diagnosis and staging. However, it is so far performed mostly manually, which is very time consuming. We formulate the detection as a multiple instance learning (MIL) problem, where instances are grouped into bags and only bag labels are available. In our case, instances are regions in the image and bags correspond to images. This has the advantage of requiring only subject-level annotation (ground truth), which is much easier to get than voxel-level manual segmentation. We consider a generalization of the standard MIL formulation where we introduce a threshold on the number of required positive instances in positive bags. This corresponds better to the classification procedure used by the radiology experts and is more robust with respect to false positive instances. We extend several existing MIL algorithms to solve the generalized case by estimating the threshold during learning. We compare the proposed methods with the baseline method on a dataset of 220 subjects. We show that the generalized MIL formulation outperforms standard MIL methods for this task. For the task of distinguishing between healthy controls and MM patients with infiltrations, our best method makes almost no mistakes with a mean AUC of 0.982 and F-1 = 0.965. We outperform the baseline method significantly in all conducted experiments.

Název v anglickém jazyce

Detecting Multiple Myeloma via Generalized Multiple-Instance Learning

Popis výsledku anglicky

We address the task of automatic detection of lesions caused by multiple myeloma (MM) in femurs or other long bones from CT data. Such detection is already an important part of the multiple myeloma diagnosis and staging. However, it is so far performed mostly manually, which is very time consuming. We formulate the detection as a multiple instance learning (MIL) problem, where instances are grouped into bags and only bag labels are available. In our case, instances are regions in the image and bags correspond to images. This has the advantage of requiring only subject-level annotation (ground truth), which is much easier to get than voxel-level manual segmentation. We consider a generalization of the standard MIL formulation where we introduce a threshold on the number of required positive instances in positive bags. This corresponds better to the classification procedure used by the radiology experts and is more robust with respect to false positive instances. We extend several existing MIL algorithms to solve the generalized case by estimating the threshold during learning. We compare the proposed methods with the baseline method on a dataset of 220 subjects. We show that the generalized MIL formulation outperforms standard MIL methods for this task. For the task of distinguishing between healthy controls and MM patients with infiltrations, our best method makes almost no mistakes with a mean AUC of 0.982 and F-1 = 0.965. We outperform the baseline method significantly in all conducted experiments.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

30224 - Radiology, nuclear medicine and medical imaging

Projekt

<a href="/cs/project/GA17-15361S" target="_blank" >GA17-15361S: Učení lokálních konceptů z globálních trénovacích dat pro klasifikaci a segmentaci biomedicínských obrazů</a><br>

Návaznosti

V - Vyzkumna aktivita podporovana z jinych verejnych zdroju

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Medical Imaging 2018: Image Processing

ISBN

978-1-5106-1638-7

ISSN

0277-786X

e-ISSN

neuvedeno

Počet stran výsledku

6

Strana od-do

—

Název nakladatele

SPIE

Místo vydání

Bellingham

Místo konání akce

Houston

Datum konání akce

11. 2. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000435027500021