Stefan Strydom
Research Subjects
Stefan Strydom has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Cristina Mannie
Johns Hopkins Bloomberg School of Public Health, Baltimore, USA
2 shared publications - Melina Taylor
Evidera Wilmington North Carolina USA
2 shared publications - Andrei Michael Dreyer
Whirly Labs, Cape Town, South Africa
1 shared publication Andrew Bazemore
American Board of Family Medicine, 1648 McGrathiana Parkway Suite 550, Lexington, KY, 40511, USA
1 shared publication- Andrew Bazemore
American Board of Family Medicine
1 shared publication - Brink van der Merwe
Stellenbosch University
1 shared publication
Latest Publications
- Open Access
- 28 April 2023
- Published by Pensoft Publishers in JUCS - Journal of Universal Computer Science
- Vol. 29 (4) , 349-373
Automatic assignment of diagnosis codes to free-form text medical note
International Classification of Disease (ICD) coding plays a significant role in classify-ing morbidity and mortality rates. Currently, ICD codes are assigned to a patient’s medical record by hand by medical practitioners or specialist clinical coders. This practice is prone to errors, and training skilled clinical coders requires time and human resources. Automatic prediction of ICD codes can help alleviate this burden. In this paper, we propose a transformer-based architecture with label-wise attention for predicting ICD codes on a medical dataset. The transformer model is first pre-trained from scratch on a medical dataset. Once this is done, the pre-trained model is used to generate representations of the tokens in the clinical documents, which are fed into the label-wise attention layer. Finally, the outputs from the label-wise attention layer are fed into a feed-forward neural network to predict appropriate ICD codes for the input document. We evaluate our model using hospital discharge summaries and their corresponding ICD-9 codes from the MIMIC-III dataset. Our experimental results show that our transformer model outperforms all previous models in terms of micro-F1 for the full label set from the MIMIC-III dataset. This is also the first successful application of a pre-trained transformer architecture to the auto-coding problem on the full MIMIC-III dataset.
- 28 April 2023
- Published by Pensoft Publishers in JUCS - Journal of Universal Computer Science
- Vol. 29 (4) , 349-373
Open AccessAutomatic assignment of diagnosis codes to free-form text medical note
International Classification of Disease (ICD) coding plays a significant role in classify-ing morbidity and mortality rates. Currently, ICD codes are assigned to a patient’s medical record by hand by medical practitioners or specialist clinical coders. This practice is prone to errors, and training skilled clinical coders requires time and human resources. Automatic prediction of ICD codes can help alleviate this burden. In this paper, we propose a transformer-based architecture with label-wise attention for predicting ICD codes on a medical dataset. The transformer model is first pre-trained from scratch on a medical dataset. Once this is done, the pre-trained model is used to generate representations of the tokens in the clinical documents, which are fed into the label-wise attention layer. Finally, the outputs from the label-wise attention layer are fed into a feed-forward neural network to predict appropriate ICD codes for the input document. We evaluate our model using hospital discharge summaries and their corresponding ICD-9 codes from the MIMIC-III dataset. Our experimental results show that our transformer model outperforms all previous models in terms of micro-F1 for the full label set from the MIMIC-III dataset. This is also the first successful application of a pre-trained transformer architecture to the auto-coding problem on the full MIMIC-III dataset.
- Open Access
- 13 January 2022
- Published by Elsevier in SSM - Qualitative Research in Health
- Vol. 2, 100041
Multinational primary health care experiences from the initial wave of the COVID-19 pandemic: A qualitative analysis
- 13 January 2022
- Published by Elsevier in SSM - Qualitative Research in Health
- Vol. 2, 100041
Open AccessMultinational primary health care experiences from the initial wave of the COVID-19 pandemic: A qualitative analysis
- Open Access
- 10 June 2021
- Published by Cambridge University Press (CUP) in Primary Health Care Research & Development
- Vol. 22, e27
Integrating primary care and public health to enhance response to a pandemic
Primary health care (PHC) includes both primary care (PC) and essential public health (PH) functions. While much is written about the need to coordinate these two aspects, successful integration remains elusive in many countries. Furthermore, the current global pandemic has highlighted many gaps in a well-integrated PHC approach. Four key actions have been recognized as important for effective integration. A survey of PC stakeholders (clinicians, researchers, and policy-makers) from 111 countries revealed many of the challenges encountered when facing the pandemic without a coordinated effort between PC and PH functions. Participants’ responses to open-ended questions underscored how each of the key actions could have been strengthened in their country and are potential factors to why a strong PC system may not have contributed to reduced mortality. By integrating PC and PH greater capacity to respond to emergencies may be possible if the synergies gained by harmonizing the two are realized.
- 10 June 2021
- Published by Cambridge University Press (CUP) in Primary Health Care Research & Development
- Vol. 22, e27
Open AccessIntegrating primary care and public health to enhance response to a pandemic
Primary health care (PHC) includes both primary care (PC) and essential public health (PH) functions. While much is written about the need to coordinate these two aspects, successful integration remains elusive in many countries. Furthermore, the current global pandemic has highlighted many gaps in a well-integrated PHC approach. Four key actions have been recognized as important for effective integration. A survey of PC stakeholders (clinicians, researchers, and policy-makers) from 111 countries revealed many of the challenges encountered when facing the pandemic without a coordinated effort between PC and PH functions. Participants’ responses to open-ended questions underscored how each of the key actions could have been strengthened in their country and are potential factors to why a strong PC system may not have contributed to reduced mortality. By integrating PC and PH greater capacity to respond to emergencies may be possible if the synergies gained by harmonizing the two are realized.
- Open Access
- 17 October 2017
- Published by Ubiquity Press, Ltd. in International Journal of Integrated Care
- Vol. 17 (5) , 322
Targeting emerging risk individuals for care management interventions
- 17 October 2017
- Published by Ubiquity Press, Ltd. in International Journal of Integrated Care
- Vol. 17 (5) , 322
Open AccessTargeting emerging risk individuals for care management interventions