Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Aug;146(11):1445-1451.
doi: 10.1017/S0950268818001498. Epub 2018 Jun 8.

Prediction of Shigellosis outcomes in Israel using machine learning classifiers

G Adamker  1 , T Holzer  1 , I Karakis  2 , M Amitay  1 , E Anis  2 , S R Singer  2 , Z Barnett-Itzhaki  1
Affiliations

Prediction of Shigellosis outcomes in Israel using machine learning classifiers

G Adamker et al. Epidemiol Infect. 2018 Aug.

Abstract

Shigellosis causes significant morbidity and mortality in developing and developed countries, mostly among infants and young children. The World Health Organization estimates that more than one million people die from Shigellosis every year. In order to evaluate trends in Shigellosis in Israel in the years 2002-2015, we analysed national notifiable disease reporting data. Shigella sonnei was the most commonly identified Shigella species in Israel. Hospitalisation rates due to Shigella flexenri were higher in comparison with other Shigella species. Shigella morbidity was higher among infants and young children (age 0-5 years old). Incidence of Shigella species differed among various ethnic groups, with significantly high rates of S. flexenri among Muslims, in comparison with Jews, Druze and Christians. In order to improve the current Shigellosis clinical diagnosis, we developed machine learning algorithms to predict the Shigella species and whether a patient will be hospitalised or not, based on available demographic and clinical data. The algorithms' performances yielded an accuracy of 93.2% (Shigella species) and 94.9% (hospitalisation) and may consequently improve the diagnosis and treatment of the disease.

Keywords: Epidemiology; Shigella; health statistics.

PubMed Disclaimer

Conflict of interest statement

None.

Figures

Fig. 1.
Fig. 1.
Analysis pipeline – Shigellosis clinical data are collected. Machine learning classifiers are trained on this data to create two models: prediction of hospitalisation and prediction of Shigella species.
Fig. 2.
Fig. 2.
Shigellosis morbidity and hospitalisation in Israel – 2002–2015. (a) Percentage of patients (out of all Shigella cases) by age groups. (b) Percentage of hospitalised patients (out of all Shigella cases) by age groups. (c) Number of cases and incidence rates of Shigellosis (2002–2015).
Fig. 3.
Fig. 3.
Distribution of Shigella species in Israel, 2002–2015 in the four major ethnic groups in Israel.
Fig. 4.
Fig. 4.
Total cases of Shigellosis morbidity sorted by month (2002–2015).
Fig. 5.
Fig. 5.
Percentage of Jewish and Muslim hospitalised patients by age groups (2002–2015).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Yang J et al. (2005) Genome dynamics and diversity of Shigella species, the etiologic agents of bacillary dysentery. Nucleic Acids Research 33, 6445–6458. - PMC - PubMed
    1. CDC. Shigella – Shigellosis. Available at https://www.cdc.gov/shigella/index.html (Accessed 14 February 2018).
    1. Ansaruzzaman M et al. (2001) Epidemiology of postshigellosis persistent diarrhea in young children. Pediatric Infectious Disease Journal 20, 525–530. - PubMed
    1. World Health Organization. Guidelines for the control of shigellosis, including epidemics due to Shigella. Available at http://apps.who.int/iris/bitstream/10665/43252/1/924159330X.pdf (Accessed 14 February 2018).
    1. Iwamoto M et al. (2014) Incidence and trends of infection with pathogens transmitted commonly through food–foodborne diseases active surveillance network, 10 U.S. Sites, 2006–2013. MMWR Morbidity and Mortality Weekly Report 63, 328–332. - PMC - PubMed

MeSH terms