Deep-learning models for differentiation of xanthogranulomatous cholecystitis and gallbladder cancer on ultrasound
© 2023. Indian Society of Gastroenterology..
BACKGROUND: The radiological differentiation of xanthogranulomatous cholecystitis (XGC) and gallbladder cancer (GBC) is challenging yet critical. We aimed at utilizing the deep learning (DL)-based approach for differentiating XGC and GBC on ultrasound (US).
METHODS: This single-center study comprised consecutive patients with XGC and GBC from a prospectively acquired database who underwent pre-operative US evaluation of the gallbladder lesions. The performance of state-of-the-art (SOTA) DL models (GBCNet-convolutional neural network [CNN] and RadFormer, transformer) for XGC vs. GBC classification in US images was tested and compared with popular DL models and a radiologist.
RESULTS: Twenty-five patients with XGC (mean age, 57 ± 12.3, 17 females) and 55 patients with GBC (mean age, 54.6 ± 11.9, 38 females) were included. The performance of GBCNet and RadFormer was comparable (sensitivity 89.1% vs. 87.3%, p = 0.738; specificity 72% vs. 84%, p = 0.563; and AUC 0.744 vs. 0.751, p = 0.514). The AUCs of DenseNet-121, vision transformer (ViT) and data-efficient image transformer (DeiT) were significantly smaller than of GBCNet (p = 0.015, 0.046, 0.013, respectively) and RadFormer (p = 0.012, 0.027, 0.007, respectively). The radiologist labeled US images of 24 (30%) patients non-diagnostic. In the remaining patients, the sensitivity, specificity and AUC for GBC detection were 92.7%, 35.7% and 0.642, respectively. The specificity of the radiologist was significantly lower than of GBCNet and RadFormer (p = 0.001).
CONCLUSION: SOTA DL models have a better performance than radiologists in differentiating XGC and GBC on the US.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2023 |
---|---|
Erschienen: |
2023 |
Enthalten in: |
Zur Gesamtaufnahme - year:2023 |
---|---|
Enthalten in: |
Indian journal of gastroenterology : official journal of the Indian Society of Gastroenterology - (2023) vom: 18. Dez. |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Gupta, Pankaj [VerfasserIn] |
---|
Links: |
---|
Themen: |
Computer |
---|
Anmerkungen: |
Date Revised 19.12.2023 published: Print-Electronic Citation Status Publisher |
---|
doi: |
10.1007/s12664-023-01483-0 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM366014811 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM366014811 | ||
003 | DE-627 | ||
005 | 20231227134619.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231227s2023 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1007/s12664-023-01483-0 |2 doi | |
028 | 5 | 2 | |a pubmed24n1232.xml |
035 | |a (DE-627)NLM366014811 | ||
035 | |a (NLM)38110782 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Gupta, Pankaj |e verfasserin |4 aut | |
245 | 1 | 0 | |a Deep-learning models for differentiation of xanthogranulomatous cholecystitis and gallbladder cancer on ultrasound |
264 | 1 | |c 2023 | |
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 Revised 19.12.2023 | ||
500 | |a published: Print-Electronic | ||
500 | |a Citation Status Publisher | ||
520 | |a © 2023. Indian Society of Gastroenterology. | ||
520 | |a BACKGROUND: The radiological differentiation of xanthogranulomatous cholecystitis (XGC) and gallbladder cancer (GBC) is challenging yet critical. We aimed at utilizing the deep learning (DL)-based approach for differentiating XGC and GBC on ultrasound (US) | ||
520 | |a METHODS: This single-center study comprised consecutive patients with XGC and GBC from a prospectively acquired database who underwent pre-operative US evaluation of the gallbladder lesions. The performance of state-of-the-art (SOTA) DL models (GBCNet-convolutional neural network [CNN] and RadFormer, transformer) for XGC vs. GBC classification in US images was tested and compared with popular DL models and a radiologist | ||
520 | |a RESULTS: Twenty-five patients with XGC (mean age, 57 ± 12.3, 17 females) and 55 patients with GBC (mean age, 54.6 ± 11.9, 38 females) were included. The performance of GBCNet and RadFormer was comparable (sensitivity 89.1% vs. 87.3%, p = 0.738; specificity 72% vs. 84%, p = 0.563; and AUC 0.744 vs. 0.751, p = 0.514). The AUCs of DenseNet-121, vision transformer (ViT) and data-efficient image transformer (DeiT) were significantly smaller than of GBCNet (p = 0.015, 0.046, 0.013, respectively) and RadFormer (p = 0.012, 0.027, 0.007, respectively). The radiologist labeled US images of 24 (30%) patients non-diagnostic. In the remaining patients, the sensitivity, specificity and AUC for GBC detection were 92.7%, 35.7% and 0.642, respectively. The specificity of the radiologist was significantly lower than of GBCNet and RadFormer (p = 0.001) | ||
520 | |a CONCLUSION: SOTA DL models have a better performance than radiologists in differentiating XGC and GBC on the US | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Computer | |
650 | 4 | |a Deep learning | |
650 | 4 | |a Gallbladder cancer | |
650 | 4 | |a Neural network | |
650 | 4 | |a Ultrasound | |
700 | 1 | |a Basu, Soumen |e verfasserin |4 aut | |
700 | 1 | |a Yadav, Thakur Deen |e verfasserin |4 aut | |
700 | 1 | |a Kaman, Lileswar |e verfasserin |4 aut | |
700 | 1 | |a Irrinki, Santosh |e verfasserin |4 aut | |
700 | 1 | |a Singh, Harjeet |e verfasserin |4 aut | |
700 | 1 | |a Prakash, Gaurav |e verfasserin |4 aut | |
700 | 1 | |a Gupta, Parikshaa |e verfasserin |4 aut | |
700 | 1 | |a Nada, Ritambhra |e verfasserin |4 aut | |
700 | 1 | |a Dutta, Usha |e verfasserin |4 aut | |
700 | 1 | |a Sandhu, Manavjit Singh |e verfasserin |4 aut | |
700 | 1 | |a Arora, Chetan |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Indian journal of gastroenterology : official journal of the Indian Society of Gastroenterology |d 1993 |g (2023) vom: 18. Dez. |w (DE-627)NLM012713031 |x 0975-0711 |7 nnns |
773 | 1 | 8 | |g year:2023 |g day:18 |g month:12 |
856 | 4 | 0 | |u http://dx.doi.org/10.1007/s12664-023-01483-0 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |j 2023 |b 18 |c 12 |