PMID- 36180775
OWN - NLM
STAT- MEDLINE
DCOM- 20221004
LR  - 20221130
IS  - 2045-2322 (Electronic)
IS  - 2045-2322 (Linking)
VI  - 12
IP  - 1
DP  - 2022 Sep 30
TI  - Accurate and fast identification of minimally prepared bacteria phenotypes using 
      Raman spectroscopy assisted by machine learning.
PG  - 16436
LID - 10.1038/s41598-022-20850-z [doi]
LID - 16436
AB  - The worldwide increase of antimicrobial resistance (AMR) is a serious threat to 
      human health. To avert the spread of AMR, fast reliable diagnostics tools that 
      facilitate optimal antibiotic stewardship are an unmet need. In this regard, 
      Raman spectroscopy promises rapid label- and culture-free identification and 
      antimicrobial susceptibility testing (AST) in a single step. However, even though 
      many Raman-based bacteria-identification and AST studies have demonstrated 
      impressive results, some shortcomings must be addressed. To bridge the gap 
      between proof-of-concept studies and clinical application, we have developed 
      machine learning techniques in combination with a novel data-augmentation 
      algorithm, for fast identification of minimally prepared bacteria phenotypes and 
      the distinctions of methicillin-resistant (MR) from methicillin-susceptible (MS) 
      bacteria. For this we have implemented a spectral transformer model for 
      hyper-spectral Raman images of bacteria. We show that our model outperforms the 
      standard convolutional neural network models on a multitude of classification 
      problems, both in terms of accuracy and in terms of training time. We attain more 
      than 96% classification accuracy on a dataset consisting of 15 different classes 
      and 95.6% classification accuracy for six MR-MS bacteria species. More 
      importantly, our results are obtained using only fast and easy-to-produce 
      training and test data.
CI  - (c) 2022. The Author(s).
FAU - Thomsen, Benjamin Lundquist
AU  - Thomsen BL
AD  - Danish Fundamental Metrology, Kogle Alle 5, 2970, Horsholm, Denmark.
FAU - Christensen, Jesper B
AU  - Christensen JB
AD  - Danish Fundamental Metrology, Kogle Alle 5, 2970, Horsholm, Denmark.
FAU - Rodenko, Olga
AU  - Rodenko O
AD  - Danish Fundamental Metrology, Kogle Alle 5, 2970, Horsholm, Denmark.
FAU - Usenov, Iskander
AU  - Usenov I
AD  - Institute of Optics and Atomic Physics, Technische Universitat Berlin, Strasse des 
      17. Juni 135, 10623, Berlin, Germany.
AD  - Art photonics GmbH, Rudower Ch 46, 12489, Berlin, Germany.
FAU - Gronnemose, Rasmus Birkholm
AU  - Gronnemose RB
AD  - Research Unit of Clinical Microbiology, University of Southern Denmark and Odense 
      University Hospital, J.B. Winslows Vej 21.2, 5000, Odense, Denmark.
FAU - Andersen, Thomas Emil
AU  - Andersen TE
AD  - Research Unit of Clinical Microbiology, University of Southern Denmark and Odense 
      University Hospital, J.B. Winslows Vej 21.2, 5000, Odense, Denmark.
FAU - Lassen, Mikael
AU  - Lassen M
AD  - Danish Fundamental Metrology, Kogle Alle 5, 2970, Horsholm, Denmark. ml@dfm.dk.
LA  - eng
PT  - Journal Article
PT  - Research Support, Non-U.S. Gov't
DEP - 20220930
PL  - England
TA  - Sci Rep
JT  - Scientific reports
JID - 101563288
RN  - 0 (Anti-Infective Agents)
RN  - Q91FH1328A (Methicillin)
SB  - IM
MH  - *Anti-Infective Agents
MH  - Bacteria
MH  - Humans
MH  - Machine Learning
MH  - Methicillin
MH  - Phenotype
MH  - *Spectrum Analysis, Raman/methods
PMC - PMC9524333
COIS- The authors declare no competing interests.
EDAT- 2022/10/01 06:00
MHDA- 2022/10/05 06:00
PMCR- 2022/09/30
CRDT- 2022/09/30 23:41
PHST- 2022/06/26 00:00 [received]
PHST- 2022/09/20 00:00 [accepted]
PHST- 2022/09/30 23:41 [entrez]
PHST- 2022/10/01 06:00 [pubmed]
PHST- 2022/10/05 06:00 [medline]
PHST- 2022/09/30 00:00 [pmc-release]
AID - 10.1038/s41598-022-20850-z [pii]
AID - 20850 [pii]
AID - 10.1038/s41598-022-20850-z [doi]
PST - epublish
SO  - Sci Rep. 2022 Sep 30;12(1):16436. doi: 10.1038/s41598-022-20850-z.