PMID- 38397962
OWN - NLM
STAT- PubMed-not-MEDLINE
LR  - 20240227
IS  - 2227-9059 (Print)
IS  - 2227-9059 (Electronic)
IS  - 2227-9059 (Linking)
VI  - 12
IP  - 2
DP  - 2024 Feb 3
TI  - Prediction of Non-Muscle Invasive Papillary Urothelial Carcinoma Relapse from 
      Hematoxylin-Eosin Images Using Deep Multiple Instance Learning in Patients 
      Treated with Bacille Calmette-Guerin Immunotherapy.
LID - 10.3390/biomedicines12020360 [doi]
LID - 360
AB  - The limited reproducibility of the grading of non-muscle invasive papillary 
      urothelial carcinoma (NMIPUC) necessitates the search for more robust image-based 
      predictive factors. In a cohort of 157 NMIPUC patients treated with Bacille 
      Calmette-Guerin (BCG) immunotherapy, we explored the multiple instance learning 
      (MIL)-based classification approach for the prediction of 2-year and 5-year 
      relapse-free survival and the multiple instance survival learning (MISL) 
      framework for survival regression. We used features extracted from image patches 
      sampled from whole slide images of hematoxylin-eosin-stained transurethral 
      resection (TUR) NPMIPUC specimens and tested several patch sampling and feature 
      extraction network variations to optimize the model performance. We selected the 
      model showing the best patient survival stratification for further testing in the 
      context of clinical and pathological variables. MISL with the multiresolution 
      patch sampling technique achieved the best patient risk stratification 
      (concordance index = 0.574, p = 0.010), followed by a 2-year MIL classification. 
      The best-selected model revealed an independent prognostic value in the context 
      of other clinical and pathologic variables (tumor stage, grade, and presence of 
      tumor on the repeated TUR) with statistically significant patient risk 
      stratification. Our findings suggest that MISL-based predictions can improve 
      NMIPUC patient risk stratification, while validation studies are needed to test 
      the generalizability of our models.
FAU - Drachneris, Julius
AU  - Drachneris J
AUID- ORCID: 0000-0003-4818-8316
AD  - Department of Pathology and Forensic Medicine, Institute of Biomedical Sciences, 
      Faculty of Medicine, Vilnius University, 03101 Vilnius, Lithuania.
AD  - National Center of Pathology, Affiliate of Vilnius University Hospital Santaros 
      Klinikos, 08406 Vilnius, Lithuania.
FAU - Morkunas, Mindaugas
AU  - Morkunas M
AUID- ORCID: 0000-0002-1360-6533
AD  - Clinic of Gastroenterology, Nephrourology and Surgery, Institute of Clinical 
      Medicine, Faculty of Medicine, Vilnius University, 08406 Vilnius, Lithuania.
FAU - Fabijonavicius, Mantas
AU  - Fabijonavicius M
AD  - Center of Urology, Vilnius University Hospital Santaros Klinikos, 08406 Vilnius, 
      Lithuania.
FAU - Cekauskas, Albertas
AU  - Cekauskas A
AD  - Clinic of Gastroenterology, Nephrourology and Surgery, Institute of Clinical 
      Medicine, Faculty of Medicine, Vilnius University, 08406 Vilnius, Lithuania.
AD  - Center of Urology, Vilnius University Hospital Santaros Klinikos, 08406 Vilnius, 
      Lithuania.
FAU - Jankevicius, Feliksas
AU  - Jankevicius F
AD  - Clinic of Gastroenterology, Nephrourology and Surgery, Institute of Clinical 
      Medicine, Faculty of Medicine, Vilnius University, 08406 Vilnius, Lithuania.
AD  - Center of Urology, Vilnius University Hospital Santaros Klinikos, 08406 Vilnius, 
      Lithuania.
FAU - Laurinavicius, Arvydas
AU  - Laurinavicius A
AUID- ORCID: 0000-0001-9232-1730
AD  - Department of Pathology and Forensic Medicine, Institute of Biomedical Sciences, 
      Faculty of Medicine, Vilnius University, 03101 Vilnius, Lithuania.
AD  - National Center of Pathology, Affiliate of Vilnius University Hospital Santaros 
      Klinikos, 08406 Vilnius, Lithuania.
LA  - eng
GR  - S-MIP-21-31/Lietuvos Mokslo Taryba/
PT  - Journal Article
DEP - 20240203
PL  - Switzerland
TA  - Biomedicines
JT  - Biomedicines
JID - 101691304
PMC - PMC10886666
OTO - NOTNLM
OT  - bladder cancer
OT  - cancer prognosis
OT  - deep learning
OT  - digital image analysis
OT  - feature extraction
OT  - survival prediction
COIS- The authors declare no conflicts of interest.
EDAT- 2024/02/24 11:45
MHDA- 2024/02/24 11:46
PMCR- 2024/02/03
CRDT- 2024/02/24 01:10
PHST- 2023/12/28 00:00 [received]
PHST- 2024/01/29 00:00 [revised]
PHST- 2024/02/01 00:00 [accepted]
PHST- 2024/02/24 11:46 [medline]
PHST- 2024/02/24 11:45 [pubmed]
PHST- 2024/02/24 01:10 [entrez]
PHST- 2024/02/03 00:00 [pmc-release]
AID - biomedicines12020360 [pii]
AID - biomedicines-12-00360 [pii]
AID - 10.3390/biomedicines12020360 [doi]
PST - epublish
SO  - Biomedicines. 2024 Feb 3;12(2):360. doi: 10.3390/biomedicines12020360.