TY - GEN
T1 - DCIS AI-TIL
T2 - 1st Workshop on Artificial Intelligence over Infrared Images for Medical Applications, AIIIMA 2022, and the 1st Workshop on Medical Image Assisted Biomarker Discovery, MIABID 2022, both held in conjunction with 25th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2022
AU - Hagos, Yeman Brhane
AU - Sobhani, Faranak
AU - Castillo, Simon P.
AU - Hall, Allison H.
AU - AbdulJabbar, Khalid
AU - Salgado, Roberto
AU - Harmon, Bryan
AU - Gallagher, Kristalyn
AU - Kilgore, Mark
AU - King, Lorraine M.
AU - Marks, Jeffrey R.
AU - Maley, Carlo
AU - Horlings, Hugo M.
AU - West, Robert
AU - Hwang, E. Shelley
AU - Yuan, Yinyin
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - Tumour infiltrating lymphocytes (TIL) influence the prognosis of Ductal carcinoma in situ (DCIS). Currently, manual assessment of TIL by expert pathologists is considered a gold standard. However, there are issues with a shortage of expert pathologists and inter-observer variability. A reliable automated scoring method is yet to be developed due to the inherent complexity of DCIS duct morphology and the assessment strategy. We developed a new deep learning and spatial analysis pipeline to automatically score DCIS stromal TIL (AI-TIL) from 243 diagnostic haematoxylin and eosin-stained whole slide images from 127 patients. To automatically identify and segment DCIS ducts, we implemented a generative adversarial network. To identify lymphocytes, we used a pre-trained deep learning model. Our DCIS segmentation model achieved a dice overlap of 0.94 (± 0.01 ) and the cell classifier model achieved 92% accuracy compared to pathologists’ annotations. Subsequently, we automatically delineated a stromal boundary and computed the percentage of the boundary area occupied by lymphocytes for each DCIS duct. Finally, we computed TIL score as the average of all duct level scores within the slide. We observe a higher correlation between AI-TIL and pathologists (average) score for wider stomal boundaries (r = 0.66, p = 6.0 × 10 - 7, W = 0.3 mm) compared with smaller boundary (r = 0.23, p = 0.12, W = 0.03 mm). Using multivariate analysis, a low AI-TIL score was associated with an increased risk of recurrence independent of age, grade, estrogen receptor (ER) status, progesterone receptor (PR) status, and necrosis (hazard ratio = 0.14, 95% CI 0.038–0.51, p = 0.003, W = 0.03 mm). These results suggest that our pipeline could be used to automatically quantify stromal TIL in DCIS and integrating AI-TIL with pathologists’ visual assessment may improve DCIS recurrence risk estimation.
AB - Tumour infiltrating lymphocytes (TIL) influence the prognosis of Ductal carcinoma in situ (DCIS). Currently, manual assessment of TIL by expert pathologists is considered a gold standard. However, there are issues with a shortage of expert pathologists and inter-observer variability. A reliable automated scoring method is yet to be developed due to the inherent complexity of DCIS duct morphology and the assessment strategy. We developed a new deep learning and spatial analysis pipeline to automatically score DCIS stromal TIL (AI-TIL) from 243 diagnostic haematoxylin and eosin-stained whole slide images from 127 patients. To automatically identify and segment DCIS ducts, we implemented a generative adversarial network. To identify lymphocytes, we used a pre-trained deep learning model. Our DCIS segmentation model achieved a dice overlap of 0.94 (± 0.01 ) and the cell classifier model achieved 92% accuracy compared to pathologists’ annotations. Subsequently, we automatically delineated a stromal boundary and computed the percentage of the boundary area occupied by lymphocytes for each DCIS duct. Finally, we computed TIL score as the average of all duct level scores within the slide. We observe a higher correlation between AI-TIL and pathologists (average) score for wider stomal boundaries (r = 0.66, p = 6.0 × 10 - 7, W = 0.3 mm) compared with smaller boundary (r = 0.23, p = 0.12, W = 0.03 mm). Using multivariate analysis, a low AI-TIL score was associated with an increased risk of recurrence independent of age, grade, estrogen receptor (ER) status, progesterone receptor (PR) status, and necrosis (hazard ratio = 0.14, 95% CI 0.038–0.51, p = 0.003, W = 0.03 mm). These results suggest that our pipeline could be used to automatically quantify stromal TIL in DCIS and integrating AI-TIL with pathologists’ visual assessment may improve DCIS recurrence risk estimation.
KW - DCIS
KW - Deep learning
KW - Tumour infiltrating lymphocyte
UR - http://www.scopus.com/inward/record.url?scp=85144823570&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85144823570&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-19660-7_16
DO - 10.1007/978-3-031-19660-7_16
M3 - Conference contribution
AN - SCOPUS:85144823570
SN - 9783031196591
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 164
EP - 175
BT - Artificial Intelligence over Infrared Images for Medical Applications and Medical Image Assisted Biomarker Discovery - 1st MICCAI Workshop, AIIIMA 2022, and 1st MICCAI Workshop, MIABID 2022, Held in Conjunction with MICCAI 2022, Proceedings
A2 - Kakileti, Siva Teja
A2 - Manjunath, Geetha
A2 - Gabrani, Maria
A2 - Rosen-Zvi, Michal
A2 - Braman, Nathaniel
A2 - Schwartz, Robert G.
A2 - Frangi, Alejandro F.
A2 - Chung, Pau-Choo
A2 - Weight, Christopher
A2 - Jagadish, Vekataraman
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 18 September 2022 through 22 September 2022
ER -