Objective. To achieve instance segmentation of upper aerodigestive tract (UADT) neoplasms using a deep learning (DL) algorithm, and to identify differences in its diagnostic performance in three different sites: larynx/hypopharynx, oral cavity and oropharynx.Methods. A total of 1034 endoscopic images from 323 patients were examined under narrow band imaging (NBI). The Mask R-CNN algorithm was used for the analysis. The dataset split was: 935 training, 48 validation and 51 testing images. Dice Similarity Coefficient (Dsc) was the main outcome measure.Results. Instance segmentation was effective in 76.5% of images. The mean Dsc was 0.90 & PLUSMN; 0.05. The algorithm correctly predicted 77.8%, 86.7% and 55.5% of lesions in the larynx/hypopharynx, oral cavity, and oropharynx, respectively. The mean Dsc was 0.90 & PLUSMN; 0.05 for the larynx/hypopharynx, 0.60 & PLUSMN; 0.26 for the oral cavity, and 0.81 & PLUSMN; 0.30 for the oropharynx. The analysis showed inferior diagnostic results in the oral cavity compared with the larynx/hypopharynx (p < 0.001). Conclusions. The study confirms the feasibility of instance segmentation of UADT using DL algorithms and shows inferior diagnostic results in the oral cavity compared with other anatomic areas.

Instance segmentation of upper aerodigestive tract cancer: site-specific outcomes

Gennarini, Francesca;Zigliani, Gabriele;Montenegro, Claudia;Sordi, Alessandra;Piazza, Cesare
2023-01-01

Abstract

Objective. To achieve instance segmentation of upper aerodigestive tract (UADT) neoplasms using a deep learning (DL) algorithm, and to identify differences in its diagnostic performance in three different sites: larynx/hypopharynx, oral cavity and oropharynx.Methods. A total of 1034 endoscopic images from 323 patients were examined under narrow band imaging (NBI). The Mask R-CNN algorithm was used for the analysis. The dataset split was: 935 training, 48 validation and 51 testing images. Dice Similarity Coefficient (Dsc) was the main outcome measure.Results. Instance segmentation was effective in 76.5% of images. The mean Dsc was 0.90 & PLUSMN; 0.05. The algorithm correctly predicted 77.8%, 86.7% and 55.5% of lesions in the larynx/hypopharynx, oral cavity, and oropharynx, respectively. The mean Dsc was 0.90 & PLUSMN; 0.05 for the larynx/hypopharynx, 0.60 & PLUSMN; 0.26 for the oral cavity, and 0.81 & PLUSMN; 0.30 for the oropharynx. The analysis showed inferior diagnostic results in the oral cavity compared with the larynx/hypopharynx (p < 0.001). Conclusions. The study confirms the feasibility of instance segmentation of UADT using DL algorithms and shows inferior diagnostic results in the oral cavity compared with other anatomic areas.
File in questo prodotto:
File Dimensione Formato  
2336-Article Text-25251-2-10-20230725 (1).pdf

accesso aperto

Tipologia: Full Text
Licenza: DRM non definito
Dimensione 336.39 kB
Formato Adobe PDF
336.39 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11379/585290
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? 0
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 2
social impact