Towards automatic pulmonary nodule management in lung cancer screening with deep learning
The introduction of lung cancer screening programs will produce an unprecedented amount of chest CT scans in the near future, which radiologists will have to read in order to decide on a patient follow-up strategy. According to the current guidelines, the workup of screen-detected nodules strongly relies on nodule size and nodule type. In this paper, we present a deep learning system based on multi-stream multi-scale convolutional networks, which automatically classifies all nodule types relevant for nodule workup. The system processes raw CT data containing a nodule without the need for any additional information such as nodule segmentation or nodule size and learns a representation of 3D data by analyzing an arbitrary number of 2D views of a given nodule. The deep learning system was trained with data from the Italian MILD screening trial and validated on an independent set of data from the Danish DLCST screening trial. We analyze the advantage of processing nodules at multiple scales with a multi-stream convolutional network architecture, and we show that the proposed deep learning system achieves performance at classifying nodule type that surpasses the one of classical machine learning approaches and is within the inter-observer variability among four experienced human observers.
Errataetall: | |
---|---|
Medienart: |
E-Artikel |
Erscheinungsjahr: |
2017 |
---|---|
Erschienen: |
2017 |
Enthalten in: |
Zur Gesamtaufnahme - volume:7 |
---|---|
Enthalten in: |
Scientific reports - 7(2017) vom: 19. Apr., Seite 46479 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Ciompi, Francesco [VerfasserIn] |
---|
Links: |
---|
Themen: |
---|
Anmerkungen: |
Date Completed 15.07.2019 Date Revised 30.03.2022 published: Electronic ErratumIn: Sci Rep. 2017 Sep 07;7:46878. - PMID 28880026 Citation Status MEDLINE |
---|
doi: |
10.1038/srep46479 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM271089350 |
---|
LEADER | 01000caa a22002652 4500 | ||
---|---|---|---|
001 | NLM271089350 | ||
003 | DE-627 | ||
005 | 20231226193825.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231224s2017 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1038/srep46479 |2 doi | |
028 | 5 | 2 | |a pubmed24n0903.xml |
035 | |a (DE-627)NLM271089350 | ||
035 | |a (NLM)28422152 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Ciompi, Francesco |e verfasserin |4 aut | |
245 | 1 | 0 | |a Towards automatic pulmonary nodule management in lung cancer screening with deep learning |
264 | 1 | |c 2017 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 15.07.2019 | ||
500 | |a Date Revised 30.03.2022 | ||
500 | |a published: Electronic | ||
500 | |a ErratumIn: Sci Rep. 2017 Sep 07;7:46878. - PMID 28880026 | ||
500 | |a Citation Status MEDLINE | ||
520 | |a The introduction of lung cancer screening programs will produce an unprecedented amount of chest CT scans in the near future, which radiologists will have to read in order to decide on a patient follow-up strategy. According to the current guidelines, the workup of screen-detected nodules strongly relies on nodule size and nodule type. In this paper, we present a deep learning system based on multi-stream multi-scale convolutional networks, which automatically classifies all nodule types relevant for nodule workup. The system processes raw CT data containing a nodule without the need for any additional information such as nodule segmentation or nodule size and learns a representation of 3D data by analyzing an arbitrary number of 2D views of a given nodule. The deep learning system was trained with data from the Italian MILD screening trial and validated on an independent set of data from the Danish DLCST screening trial. We analyze the advantage of processing nodules at multiple scales with a multi-stream convolutional network architecture, and we show that the proposed deep learning system achieves performance at classifying nodule type that surpasses the one of classical machine learning approaches and is within the inter-observer variability among four experienced human observers | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Research Support, Non-U.S. Gov't | |
700 | 1 | |a Chung, Kaman |e verfasserin |4 aut | |
700 | 1 | |a van Riel, Sarah J |e verfasserin |4 aut | |
700 | 1 | |a Setio, Arnaud Arindra Adiyoso |e verfasserin |4 aut | |
700 | 1 | |a Gerke, Paul K |e verfasserin |4 aut | |
700 | 1 | |a Jacobs, Colin |e verfasserin |4 aut | |
700 | 1 | |a Scholten, Ernst Th |e verfasserin |4 aut | |
700 | 1 | |a Schaefer-Prokop, Cornelia |e verfasserin |4 aut | |
700 | 1 | |a Wille, Mathilde M W |e verfasserin |4 aut | |
700 | 1 | |a Marchianò, Alfonso |e verfasserin |4 aut | |
700 | 1 | |a Pastorino, Ugo |e verfasserin |4 aut | |
700 | 1 | |a Prokop, Mathias |e verfasserin |4 aut | |
700 | 1 | |a van Ginneken, Bram |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Scientific reports |d 2011 |g 7(2017) vom: 19. Apr., Seite 46479 |w (DE-627)NLM215703936 |x 2045-2322 |7 nnns |
773 | 1 | 8 | |g volume:7 |g year:2017 |g day:19 |g month:04 |g pages:46479 |
856 | 4 | 0 | |u http://dx.doi.org/10.1038/srep46479 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 7 |j 2017 |b 19 |c 04 |h 46479 |