rmnldwg/lymph
Python package for statistical modelling of lymphatic metastatic spread in head & neck cancer.
LyProX is designed to collect, share, and visualize data on lymphatic metastases in head and neck squamous cell carcinomas. By providing detailed insights into progression patterns, it aims to support researchers and clinicians in working towards more personalized elective nodal treatments.
Current clinical guidelines (e.g. by Biau et al. ) for the elective nodal clinical target volume (CTV-N) in head and neck cancer are mostly based on the prevalence of involvement in any given lymph node level (LNL). However, patients with vastly different states of lymphatic spread are often treated the same way.
The basis for a more personalized approach is the quantification of individual risk for lymphatic metastases. This, in turn, requires a comprehensive understanding of the detailed patterns of lymphatic spread. LyProX aims to provide this understanding by collecting and visualizing such data.
With an intuitive interface, LyProX allows users to explore datasets of lymphatic progression patterns interactively. For each patient in the different cohorts, it stores in which LNLs metastases were found using different diagnostic modalities or pathological assessments after a neck dissection.
This detailed per-patient and per-level information can then be filtered and aggregated in the dashboard . For example, one may be interested in the correlation between contralateral involvement and the lateralization of the primary tumor. Or how much more likely downstream levels are to be involved if upstream levels are affected.
Making the data interactively explorable fosters hypothesis generation and encourages further research into lymphatic spread patterns. Also, the data may be downloaded as a CSV table or fetched from our data repository for further analysis.
After pulishing our first datasets, we started working on probabilistic models of lymphatic tumor progression These interpretable hidden Markov models (HMMs) are trained on the data published in LyProX and accurately predict the risk of occult disease in any LNL, given a new patient's diagnosis.
rmnldwg/lymph
lycosystem/lyprox
rmnldwg/lydata
rmnldwg/lyscripts
lycosystem/lymixture
Radiotherapy and Oncology 153 (2020)
J. Unkelbach,
T. Bortfeld,
C. Cardenas,
V. Gregoire,
W. Hager,
B. Heijmen,
R. Jeraj,
S. Korreman,
R. Ludwig,
B. Pouymayou,
N. Shusharina,
J. Söderberg,
I. Toma-Dasu,
E. Troost,
E. Vasquez Osorio
The role of computational methods for automating and improving clinical target volume definition
Scientific Reports 11 (2021)
R. Ludwig,
B. Pouymayou,
P. Balermpas,
J. Unkelbach
A hidden Markov model for lymphatic tumor progression in the head and neck
Radiotherapy and Oncology 169 (2022)
R. Ludwig,
J. Hoffmann,
B. Pouymayou,
M. Däppen,
G. Morand,
M. Guckenberger,
V. Grégoire,
P. Balermpas,
J. Unkelbach
Detailed patient-individual reporting of lymph node involvement in oropharyngeal squamous cell carcinoma with an online interface
Data in Brief 43 (2022)
R. Ludwig,
J. Hoffmann,
B. Pouymayou,
G. Morand,
M. Däppen,
M. Guckenberger,
V. Grégoire,
P. Balermpas,
J. Unkelbach
A dataset on patient-individual lymph node involvement in oropharyngeal squamous cell carcinoma
Data in Brief 52 (2024)
R. Ludwig,
A. Schubert,
D. Barbatei,
L. Bauwens,
S. Werlen,
O. Elicin,
M. Dettmer,
P. Zrounba,
P. Balermpas,
B. Pouymayou,
V. Grégoire,
R. Giger,
J. Unkelbach
A multi-centric dataset on patient-individual pathological lymph node involvement in head and neck squamous cell carcinoma
Scientific Reports 14 (2024)
R. Ludwig,
A. Schubert,
D. Barbatei,
L. Bauwens,
J. Hoffmann,
S. Werlen,
O. Elicin,
M. Dettmer,
P. Zrounba,
B. Pouymayou,
P. Balermpas,
V. Grégoire,
R. Giger,
J. Unkelbach
Modelling the lymphatic metastatic progression pathways of OPSCC from multi-institutional datasets
Data in Brief 60 (2025)
S. Benavente,
R. Ludwig,
P. Balermpas,
J. Unkelbach
Updating “A dataset on patient-individual lymph node involvement in oropharyngeal squamous cell carcinoma” with an additional dataset from a second institution