Mapping microstructural features of pathological tissues by pixel clustering of Mueller matrix images
Abstract In histopathology, doctors identify diseases by characterizing abnormal cells and their spatial organization within tissues. Polarization microscopy and supervised learning have been proved as an effective tool for extracting polarization parameters to highlight pathological features. Here we present an alternative approach based on unsupervised learning to group polarization-pixels into clusters, which correspond to distinct pathological structures. For pathological samples from different patients, it is confirmed that such unsupervised learning technique can decompose the histological structures into a stable basis of characteristic microstructural clusters, some of which correspond to distinctive pathological features for clinical diagnosis. Using hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) samples, we demonstrate how the proposed framework can be utilized for segmentation of histological image, visualization of microstructure composition associated with lesion, and identification of polarization-based microstructure markers that correlates with specific pathology variation. This technique is capable of unraveling invisible microstructures in non-polarization images, and turn them into visible polarization features to pathologists and researchers..
| Medienart: |
|
|---|
Preprint| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
ResearchSquare.com - (2023) vom: 09. Dez. Zur Gesamtaufnahme - year:2023 |
|---|
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Ma, Hui [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
|---|
| doi: |
10.21203/rs.3.rs-2483307/v1 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
XRA038910454 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | XRA038910454 | ||
| 003 | DE-627 | ||
| 005 | 20231210130645.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 230310s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.21203/rs.3.rs-2483307/v1 |2 doi | |
| 035 | |a (DE-627)XRA038910454 | ||
| 035 | |a (ResearchSquare)10.21203/rs.3.rs-2483307/v1 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Ma, Hui |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Mapping microstructural features of pathological tissues by pixel clustering of Mueller matrix images |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 520 | |a Abstract In histopathology, doctors identify diseases by characterizing abnormal cells and their spatial organization within tissues. Polarization microscopy and supervised learning have been proved as an effective tool for extracting polarization parameters to highlight pathological features. Here we present an alternative approach based on unsupervised learning to group polarization-pixels into clusters, which correspond to distinct pathological structures. For pathological samples from different patients, it is confirmed that such unsupervised learning technique can decompose the histological structures into a stable basis of characteristic microstructural clusters, some of which correspond to distinctive pathological features for clinical diagnosis. Using hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) samples, we demonstrate how the proposed framework can be utilized for segmentation of histological image, visualization of microstructure composition associated with lesion, and identification of polarization-based microstructure markers that correlates with specific pathology variation. This technique is capable of unraveling invisible microstructures in non-polarization images, and turn them into visible polarization features to pathologists and researchers. | ||
| 650 | 4 | |a Biology |7 (dpeaa)DE-84 | |
| 650 | 4 | |a 570 |7 (dpeaa)DE-84 | |
| 700 | 1 | |a Wan, Jiachen |4 aut | |
| 700 | 1 | |a Dong, Yang |4 aut | |
| 700 | 1 | |a Yao, Yue |4 aut | |
| 700 | 1 | |a Xiao, Weijin |4 aut | |
| 700 | 1 | |a Huang, Ruqi |0 (orcid)0000-0001-5942-3671 |4 aut | |
| 700 | 1 | |a Xue, Jing-Hao |4 aut | |
| 700 | 1 | |a Peng, Ran |4 aut | |
| 700 | 1 | |a Pei, Haojie |4 aut | |
| 700 | 1 | |a Tian, Xuewu |4 aut | |
| 700 | 1 | |a Liao, Ran |4 aut | |
| 700 | 1 | |a He, Honghui |0 (orcid)0000-0001-7369-7433 |4 aut | |
| 700 | 1 | |a Zeng, Nan |4 aut | |
| 700 | 1 | |a Li, Chao |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t ResearchSquare.com |g (2023) vom: 09. Dez. |
| 773 | 1 | 8 | |g year:2023 |g day:09 |g month:12 |
| 856 | 4 | 0 | |u https://doi.org/10.1038/s44172-023-00136-6 |z lizenzpflichtig |3 Volltext |
| 856 | 4 | 0 | |u http://dx.doi.org/10.21203/rs.3.rs-2483307/v1 |z kostenfrei |3 Volltext |
| 912 | |a GBV_XRA | ||
| 951 | |a AR | ||
| 952 | |j 2023 |b 09 |c 12 | ||