Armita Zarnegar

Armita Zarnegar

Auto-generated profile, based on publicly available information

Armita Zarnegar

Swinburne University of Technology, Melbourne, Victoria, Australia

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 April 2022 in Journal of Clinical Epidemiology
Verified
Journal Article
Closed access
HKHanan Khalil
Hanan Khalil
DADaniel Ameen
Daniel Ameen
AZArmita Zarnegar
Armita Zarnegar

20 September 2021
Verified
Book Chapter
Open access
PDF file
BOBahadorreza Ofoghi
Bahadorreza Ofoghi
AZArmita Zarnegar
Armita Zarnegar
Question Answering (QA) systems play an important role in decision support systems. Deep neural network-based passage rankers have recently been developed to more effectively rank likely answer-containing passages for QA purposes. These rankers utilize distributed word or sentence embeddings. Such distributed representations mostly carry semantic relatedness of text units in which explicit linguistic features are under-represented. In this paper, we take novel approaches to combine linguistic features (such as different part-of-speech measures) with distributed sentence representations of questions and passages. The QUASAR-T fact-seeking questions and short text passages were used in our experiments to show that while ensembling of deep relevance measures based on pure sentence embedding with linguistic features using several machine learning techniques fails to improve upon the passage ranking performance of our baseline neural network ranker, the concatenation of the same features within the network structure significantly improves the overall performance of passage ranking for QA.

16 September 2025
Verified
Conference Paper
Open access
AZArmita Zarnegar
Armita Zarnegar
Artificial Intelligence (AI) is revolutionising healthcare by enhancing diagnostics, enabling personalised treatment, and improving patient outcome predictions. However, these systems also pose significant risks of perpetuating and amplifying biases, which could lead to inequitable healthcare delivery. This paper presents a synthesis of 25 peer-reviewed studies, narrowed down from an initial pool of 498 using eight screening criteria, focused on identifying sources of bias and evaluating mitigation strategies in medical AI. Sources of bias include imbalanced training data, algorithmic design flaws, deployment bias, and clinical implementation inequities, which often disproportionately impact marginalised populations. Notably, 22 out of the 25 studies emphasised data bias as the primary source, underscoring a strong research focus on dataset representativeness and quality. Common bias types identified include racial, gender, age-related, and socio-economic disparities. Mitigation approaches encompass pre-processing and data-centric techniques (e.g., resampling, disaggregation, data augmentation using GAN), model-centric strategies (e.g., multi-calibration and smoothing), fairness evaluation and transparency monitoring (e.g., SHAP, LIME, fairness metrics), and participatory approaches that promote transparency and stakeholder engagement. Notable gaps in the literature include limited attention to temporal fairness, mental health AI, and paediatric or geriatric specific biases. Among the selected studies, medical imaging emerged as the most frequently addressed application area. Some of these techniques, reported in single studies, indicate a need for further validation and integration into standard practice. These findings underscore the need for more comprehensive, equitable, and domain-specific approaches to responsible AI implementation in healthcare.

17 February 2026
Verified
Conference Paper
Open access
JLJack Li
Jack Li
YPYin Paradies
Yin Paradies
TMTrina Myers
Trina Myers
RDRobin Doss
Robin Doss
AZArmita Zarnegar
Armita Zarnegar
JRJack Reis
Jack Reis

16 July 2022 in JMIR Preprints
Verified
Preprint
Open access
AZArmita Zarnegar
Armita Zarnegar
UNSTRUCTURED Point of care (POC) devices enable medical interventions, including testing, to be performed near the patient and often by the patient. Some have suggested that this will disrupt medicine from a symptom/diagnosis/treatment endeavour to a monitoring/prediction/prevention exercise. The rate of innovation and discovery in this area is high, with a majority of these devices used for screening. This article reviews the public health benefits of point of care devices in terms of patient engagement, health inequality and health promotion. Low-cost screening clinics that make use of point of care devices ultimately help with reducing health inequality in developing countries and remote areas. The article also outlines the issues hindering the benefits of using POC and further suggests a framework to present and evaluate point of care devices used in the healthcare industry. The scope of this study is limited to medical-grade devices used at home, in clinics, aged care and hospital settings. We argue that having such a framework provides a neutral source of information to evaluate and select POC for various tasks. CONCLUSION This study briefly reviews the benefits and challenges associated with using emerging point of care devices to achieve public health benefits. Emerging Point of Care devices has the potential to disrupt public health by increasing patient engagement in managing their own conditions, reducing inequalities in access to healthcare caused by geographic or resource factors, the automated detection of undiagnosed conditions, provision of the healthcare system, and the assembly of large datasets that can be analysed to guide public health policy. However, for these public health benefits to be realised, POC devices need to be regulated in a more open and transparent manner. A framework is proposed to provide information in a transparent manner for the comparison, maintenance and best use of POIC devices in order to facilitate the inclusion of these devices in the health care system.

13 June 2025
Verified
Conference Paper
Open access
CLCeline Latulipe
Celine Latulipe
SMSteph McIntyre
Steph McIntyre
JAJohn Anvik
John Anvik
KLKevin Lin
Kevin Lin
SNSabrin Nowrin
Sabrin Nowrin
AZArmita Zarnegar
Armita Zarnegar