Unsupervised segmentation in NSCLC: how to map the output of unsupervised segmentation to meaningful histological labels by linear combination?
Background: Segmentation is, in many Pathomics projects, an initial step. Usually, in supervised settings, well-annotated and large datasets are required. Regarding the rarity of such datasets, unsupervised learning concepts appear to be a potential solution. Against this background, we tested for a...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
7 April 2022
|
| In: |
Applied Sciences
Year: 2022, Jahrgang: 12, Heft: 8, Pages: 1-18 |
| ISSN: | 2076-3417 |
| DOI: | 10.3390/app12083718 |
| Online-Zugang: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.3390/app12083718 Verlag, lizenzpflichtig, Volltext: https://www.mdpi.com/2076-3417/12/8/3718 
|
| Verfasserangaben: | Cleo-Aron Weis, Kian R. Weihrauch, Katharina Kriegsmann and Mark Kriegsmann |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1805154044 | ||
| 003 | DE-627 | ||
| 005 | 20220820192339.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 220530s2022 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.3390/app12083718 |2 doi | |
| 035 | |a (DE-627)1805154044 | ||
| 035 | |a (DE-599)KXP1805154044 | ||
| 035 | |a (OCoLC)1341460600 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 33 |2 sdnb | ||
| 100 | 1 | |a Weis, Cleo-Aron Thias |d 1985- |e VerfasserIn |0 (DE-588)1062803752 |0 (DE-627)806961198 |0 (DE-576)42022730X |4 aut | |
| 245 | 1 | 0 | |a Unsupervised segmentation in NSCLC |b how to map the output of unsupervised segmentation to meaningful histological labels by linear combination? |c Cleo-Aron Weis, Kian R. Weihrauch, Katharina Kriegsmann and Mark Kriegsmann |
| 246 | 3 | 0 | |a Artificial intelligence applied to medical imaging and computational biology |
| 264 | 1 | |c 7 April 2022 | |
| 300 | |a 18 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a This article belongs to the Special issue "Artificial Intelligence Applied to Medical Imaging and Computational Biology" | ||
| 500 | |a Gesehen am 30.05.2022 | ||
| 520 | |a Background: Segmentation is, in many Pathomics projects, an initial step. Usually, in supervised settings, well-annotated and large datasets are required. Regarding the rarity of such datasets, unsupervised learning concepts appear to be a potential solution. Against this background, we tested for a small dataset on lung cancer tissue microarrays (TMA) if a model (i) first can be in a previously published unsupervised setting and (ii) secondly can be modified and retrained to produce meaningful labels, and (iii) we finally compared this approach to standard segmentation models. Methods: (ad i) First, a convolutional neuronal network (CNN) segmentation model is trained in an unsupervised fashion, as recently described by Kanezaki et al. (ad ii) Second, the model is modified by adding a remapping block and is retrained on an annotated dataset in a supervised setting. (ad iii) Third, the segmentation results are compared to standard segmentation models trained on the same dataset. Results: (ad i-ii) By adding an additional mapping-block layer and by retraining, models previously trained in an unsupervised manner can produce meaningful labels. (ad iii) The segmentation quality is inferior to standard segmentation models trained on the same dataset. Conclusions: Unsupervised training in combination with subsequent supervised training offers for histological images here no benefit. | ||
| 650 | 4 | |a histopathology | |
| 650 | 4 | |a lung cancer | |
| 650 | 4 | |a supervised segmentation | |
| 650 | 4 | |a unsupervised segmentation | |
| 700 | 1 | |a Weihrauch, Kian R. |e VerfasserIn |4 aut | |
| 700 | 1 | |a Kriegsmann, Katharina |d 1986- |e VerfasserIn |0 (DE-588)1049422449 |0 (DE-627)781924006 |0 (DE-576)40339774X |4 aut | |
| 700 | 1 | |a Kriegsmann, Mark |d 1987- |e VerfasserIn |0 (DE-588)103740324X |0 (DE-627)755668782 |0 (DE-576)39141870X |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Applied Sciences |d Basel : MDPI, 2011 |g 12(2022), 8, Special issue, Artikel-ID 3718, Seite 1-18 |h Online-Ressource |w (DE-627)737287640 |w (DE-600)2704225-X |w (DE-576)379466716 |x 2076-3417 |7 nnas |a Unsupervised segmentation in NSCLC how to map the output of unsupervised segmentation to meaningful histological labels by linear combination? |
| 773 | 1 | 8 | |g volume:12 |g year:2022 |g number:8 |g month:4- |g supplement:Special issue |g elocationid:3718 |g pages:1-18 |g extent:18 |a Unsupervised segmentation in NSCLC how to map the output of unsupervised segmentation to meaningful histological labels by linear combination? |
| 856 | 4 | 0 | |u https://doi.org/10.3390/app12083718 |x Verlag |x Resolving-System |z lizenzpflichtig |3 Volltext |
| 856 | 4 | 0 | |u https://www.mdpi.com/2076-3417/12/8/3718 |x Verlag |z lizenzpflichtig |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20220530 | ||
| 993 | |a Article | ||
| 994 | |a 2022 | ||
| 998 | |g 103740324X |a Kriegsmann, Mark |m 103740324X:Kriegsmann, Mark |d 910000 |d 912000 |d 50000 |e 910000PK103740324X |e 912000PK103740324X |e 50000PK103740324X |k 0/910000/ |k 1/910000/912000/ |k 0/50000/ |p 4 |y j | ||
| 998 | |g 1049422449 |a Kriegsmann, Katharina |m 1049422449:Kriegsmann, Katharina |d 910000 |d 910100 |e 910000PK1049422449 |e 910100PK1049422449 |k 0/910000/ |k 1/910000/910100/ |p 3 | ||
| 998 | |g 1062803752 |a Weis, Cleo-Aron Thias |m 1062803752:Weis, Cleo-Aron Thias |d 60000 |d 63400 |e 60000PW1062803752 |e 63400PW1062803752 |k 0/60000/ |k 1/60000/63400/ |p 1 |x j | ||
| 999 | |a KXP-PPN1805154044 |e 4141341703 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"id":{"doi":["10.3390/app12083718"],"eki":["1805154044"]},"relHost":[{"id":{"eki":["737287640"],"issn":["2076-3417"],"zdb":["2704225-X"]},"physDesc":[{"extent":"Online-Ressource"}],"pubHistory":["1.2011 -"],"recId":"737287640","origin":[{"dateIssuedKey":"2011","publisher":"MDPI","dateIssuedDisp":"2011-","publisherPlace":"Basel"}],"note":["Gesehen am 19.02.13"],"type":{"bibl":"periodical","media":"Online-Ressource"},"title":[{"title":"Applied Sciences","title_sort":"Applied Sciences","subtitle":"open access journal"}],"part":{"volume":"12","extent":"18","text":"12(2022), 8, Special issue, Artikel-ID 3718, Seite 1-18","pages":"1-18","year":"2022","issue":"8"},"language":["eng"],"disp":"Unsupervised segmentation in NSCLC how to map the output of unsupervised segmentation to meaningful histological labels by linear combination?Applied Sciences"}],"physDesc":[{"extent":"18 S."}],"name":{"displayForm":["Cleo-Aron Weis, Kian R. Weihrauch, Katharina Kriegsmann and Mark Kriegsmann"]},"recId":"1805154044","origin":[{"dateIssuedKey":"2022","dateIssuedDisp":"7 April 2022"}],"note":["This article belongs to the Special issue \"Artificial Intelligence Applied to Medical Imaging and Computational Biology\"","Gesehen am 30.05.2022"],"title":[{"title":"Unsupervised segmentation in NSCLC","title_sort":"Unsupervised segmentation in NSCLC","subtitle":"how to map the output of unsupervised segmentation to meaningful histological labels by linear combination?"}],"type":{"media":"Online-Ressource","bibl":"article-journal"},"language":["eng"],"person":[{"family":"Weis","display":"Weis, Cleo-Aron Thias","given":"Cleo-Aron Thias","role":"aut"},{"display":"Weihrauch, Kian R.","family":"Weihrauch","role":"aut","given":"Kian R."},{"role":"aut","given":"Katharina","display":"Kriegsmann, Katharina","family":"Kriegsmann"},{"role":"aut","given":"Mark","display":"Kriegsmann, Mark","family":"Kriegsmann"}]} | ||
| SRT | |a WEISCLEOARUNSUPERVIS7202 | ||