Subscribe now to Knowmad Alert
Specialized Content Curation | February 16-22, 2021| Week 13
Healthcare education and artificial intelligence converging: "Does technology care?"
Technologies such as Deep Learning Artificial Intelligence (AI) promise benign solutions to complex problems, but for the authors of this study (A.G. van der Niet, A. Bleakleyesta) this view is misguided.
The role of AI in medicine promises to redouble efficiency by reducing complex patient experiences to linear problem-solving interventions promised by "solutionism." As an instrumental intervention, AI can objectify patients, the study notes, this can frustrate the benefits of dialogue as patients' complex and often unpredictable carnal experiences of disease are recalculated in solution-focused computational terms.
The paper explores how AI is often promoted as a holistic response to complex problems, including pedagogy, where learning "hands-on" bedside medicine has demonstrated benefits beyond the technical. Medical education faces a pedagogical challenge:
What will the physician of the future need to know that AI does not, and what will physicians continue to offer patients beyond diagnostic accuracy? In conclusion, there is a danger of falling into the trap of instrumentalism, where the curriculum itself becomes an algorithm. Students need to understand that these technologies will come to mediate and frame their perceptions of medical problems.
The authors noted that medical education has been slow to respond to the demands of early learning of AI in undergraduate medicine, so they conclude that engagement with AI will become a key factor in constructing the identity of future physicians, especially in the practice of ethical medicine.
Click here to read full paper
Niet, A. G., & Bleakley, A. (2020). Where medical education meets artificial intelligence: “Does technology care?” Medical Education, 55(1), 30–36. https://doi.org/10.1111/medu.14131
Subscribe now to Knowmad Alert
Different disease detection and classification techniques using Deep Learning for the Cannabis plant.
Cannabis plants are used for medical and recreational purposes, which is why this study proposes a classification and analysis of different disease detection and classification models for cannabis plants. Emerging technologies can help cannabis farmers facing problems in a crop that is susceptible to multiple disorders.
The models used in this study are Fast Region Convolutional Neural Network(F-RCNN), MobileNet Single Shot Multibox Detector(MobileNet-SSD), You Only Look Once(YOLO), and Residual Network-50 Layers (ResNet50). The authors found that MobileNet-SSD provides the best accuracy among all the object detection models studied and also requires the shortest training time.
During the experiments, it was concluded that MobileNet-SSD has achieved 85% accuracy for the five disease categories with three stages each. ResNet 50 shows an accuracy of 88% for detecting and classifying two different diseases with three stages each and for the five classes with three stages each, the accuracy is approximately 62%.
The accuracy of F-RCNN was lower than that of MobileNetSSD. The authors propose that while the actual dataset is not yet fully labeled for testing in object detection models, the results of the experiments instill confidence that disease detection and localization can be performed using the parameters set in the object detection models. They, therefore, propose to incorporate the structured component of the dataset to augment the detection and classification process.
Click here to read full paper
Pathak, K., Arya, A., Hatti, P., Handragal, V., & Lee, K. (2021). A Study of Different Disease Detection and Classification Techniques using Deep Learning for Cannabis Plant. International Journal of Computing and Digital Systems, 10(1), 2210–2142. https://doi.org/10.12785/ijcds/100106
Subscribe now to Knowmad Alert
Biomedical Waste from India during the COVID-19 crisis: Associated environmental-health impacts and mitigation measures.
The situations induced by the COVID-19 health crisis have strained the biomedical waste management (BMW) system globally. The sudden influx of SARS-CoV-2 infected patients in healthcare facilities has increased the generation of yellow category biomedical waste (Y-BMW) and placed a considerable burden on the biomedical waste incineration units in India. The authors of the study present the compromised situation of India's BMW incineration units in the wake of the pandemic from March 21, 2020 to August 31, 2020.
(Biomedical Waste (Management & Handling) Rules, 2011: An overview, 2011)
The sudden influx of SARS-CoV-2 infected patients in healthcare facilities has increased the generation of yellow category biomedical waste (Y-BMW) and placed a considerable burden on India's biomedical waste incineration units.
The paper notes that it was observed that on July 13, 2020, the total Y-BMW, generated by both normal and SARS-CoV-2 infected patients, fully utilized India's BMW incineration capacity. The authors highlight that during the study period, BMW incineration emitted multiple pollutants and their concentration was in the order: NOx > CO > CO > SOx > PM > HCl > Cd > Pb > Hg > PCBs > Ni > Cr > Be > As. The study shows a lifetime cancer risk assessment with hazard quotient >10-6, Cd can induce carcinogenic health effects in both adults and children in India. The results of this study revealed that, in India, a COVID-19 infected patient generates approximately 3.41 kg/d of BMW.
*Click here to read full paper
Thind, P. S., Sareen, A., Singh, D. D., Singh, S., & John, S. (2021). Compromising situation of India’s bio-medical waste incineration units during pandemic outbreak of COVID-19: Associated environmental-health impacts and mitigation measures. Environmental Pollution, 276, 116621. https://doi.org/10.1016/j.envpol.2021.116621
Subscribe now to Knowmad Alert