Development and testing of a multi-lingual Natural Language Processing-based deep learning system in 10 languages for COVID-19 pandemic crisis: A multi-center study.

Purpose: The COVID-19 pandemic has drastically disrupted global healthcare systems. With the higher demand for healthcare and misinformation related to COVID-19, there is a need to explore alternative models to improve communication. Artificial Intelligence (AI) and Natural Language Processing (NLP) have emerged as promising solutions to improve healthcare delivery. Chatbots could fill a pivotal role in the dissemination and easy accessibility of accurate information in a pandemic. In this study, we developed a multi-lingual NLP-based AI chatbot, DR-COVID, which responds accurately to open-ended, COVID-19 related questions. This was used to facilitate pandemic education and healthcare delivery. Methods: First, we developed DR-COVID with an ensemble NLP model on the Telegram platform (https://t.me/drcovid_nlp_chatbot). Second, we evaluated various performance metrics. Third, we evaluated multi-lingual text-to-text translation to Chinese, Malay, Tamil, Filipino, Thai, Japanese, French, Spanish, and Portuguese. We utilized 2,728 training questions and 821 test questions in English. Primary outcome measurements were (A) overall and top 3 accuracies; (B) Area Under the Curve (AUC), precision, recall, and F1 score. Overall accuracy referred to a correct response for the top answer, whereas top 3 accuracy referred to an appropriate response for any one answer amongst the top 3 answers. AUC and its relevant matrices were obtained from the Receiver Operation Characteristics (ROC) curve. Secondary outcomes were (A) multi-lingual accuracy; (B) comparison to enterprise-grade chatbot systems. The sharing of training and testing datasets on an open-source platform will also contribute to existing data. Results: Our NLP model, utilizing the ensemble architecture, achieved overall and top 3 accuracies of 0.838 [95% confidence interval (CI): 0.826-0.851] and 0.922 [95% CI: 0.913-0.932] respectively. For overall and top 3 results, AUC scores of 0.917 [95% CI: 0.911-0.925] and 0.960 [95% CI: 0.955-0.964] were achieved respectively. We achieved multi-linguicism with nine non-English languages, with Portuguese performing the best overall at 0.900. Lastly, DR-COVID generated answers more accurately and quickly than other chatbots, within 1.12-2.15 s across three devices tested. Conclusion: DR-COVID is a clinically effective NLP-based conversational AI chatbot, and a promising solution for healthcare delivery in the pandemic era.

Centration and Stability of Small-Aperture Intraocular Lens in Aberrated Eyes.

PURPOSE: To investigate the visual outcomes and postoperative stability of IC-8 intraocular lens (AcuFocus, Inc) implantation following femtosecond laser-assisted cataract surgery. METHODS: In this prospective study, the IC-8 IOL was implanted in the non-dominant eye of 12 patients. Centration and uncorrected and corrected distance, intermediate, and near visual acuities (UDVA, UIVA, UNVA, CDVA, CIVA, CNVA) were evaluated up to postoperative month 3 (POM3). Means ± standard deviations are reported in logMAR units. RESULTS: There were no significant differences in absolute, vertical, and horizontal centration postoperatively versus intraoperatively. From intraoperative to POM3, mean displacement was 0.17 ± 0.09 and 0.14 ± 0.07 mm (P = .52) relative to the capsulorhexis and 0.28 ± 0.22 and 0.25 ± 0.14 mm relative to the limbus (P = .62), respectively. Centration remained within the 1.36-mm IC-8 aperture. Excluding patients with ocular comorbidities, from preoperatively to POM3, UDVA improved significantly from 0.52 ± 0.39 to 0.20 ± 0.11 logMAR (P = .024), with an efficacy index of 0.80 ± 0.31. The safety index was 1.30 ± 0.40, with no adverse events experienced. At POM3, CDVA, UIVA, and UNVA of 0.04 ± 0.05, 0.27 ± 0.11, and 0.22 ± 0.10 logMAR were achieved, respectively. Binocularly, all achieved UDVA of 20/25 or better and UIVA and UNVA of N6 and better. CONCLUSIONS: The IC-8 IOL can attain good centration and positional stability up to POM3. It enables both extended depth of focus and tolerance to aberration, making it capable of achieving spectacle independence after surgery. [J Refract Surg. 2022;38(2):98-105.].

Virtual care to increase military medical centre capacity in the primary health care setting: A prospective self-controlled pilot study of symptoms collection and telemedicine.

INTRODUCTION: The Singapore Armed Forces (SAF) recognises the potential benefits and looks to harnessing telemedicine for primary health care services. In this prospective self-controlled pilot study, we aimed to evaluate the safety, efficiency and user satisfaction outcomes of virtual care (VC) at a military medical centre. METHODS: Out of 320 patients seen during the study period, 28 were enrolled in this study and underwent on-premises VC, comprising digital symptoms collection and telemedicine in addition to the usual in-person physician consultation. Safety outcomes were measured based on the diagnostic concordance between physicians. Efficiency was measured based on consultation times, and user satisfaction was evaluated using a standard questionnaire. RESULTS: There was a higher caseload of both upper respiratory infections and dermatological conditions in our population, in which telemedicine performed well. In terms of safety, telemedicine achieved a mean diagnostic concordance of 92.8% compared to in-person consultations. In terms of efficiency, consultation times were 26.2% - or 2 minutes and 15 seconds - shorter on average with telemedicine (p = 0.0488). User satisfaction was favourable, with 85.5% of patients satisfied with the VC experience. DISCUSSION: This study has been invaluable in showing that on-premises telemedicine is a safe, efficient and effective means to extend and increase our surge capacity for primary health care. Our results have given us reasonable confidence to explore a larger-scale implementation in our network of military medical centres in the future.

Deep learning-based natural language processing in ophthalmology: applications, challenges and future directions.

PURPOSE OF REVIEW: Artificial intelligence (AI) is the fourth industrial revolution in mankind's history. Natural language processing (NLP) is a type of AI that transforms human language, to one that computers can interpret and process. NLP is still in the formative stages of development in healthcare, with promising applications and potential challenges in its applications. This review provides an overview of AI-based NLP, its applications in healthcare and ophthalmology, next-generation use case, as well as potential challenges in deployment. RECENT FINDINGS: The integration of AI-based NLP systems into existing clinical care shows considerable promise in disease screening, risk stratification, and treatment monitoring, amongst others. Stakeholder collaboration, greater public acceptance, and advancing technologies will continue to shape the NLP landscape in healthcare and ophthalmology. SUMMARY: Healthcare has always endeavored to be patient centric and personalized. For AI-based NLP systems to become an eventual reality in larger-scale applications, it is pertinent for key stakeholders to collaborate and address potential challenges in application. Ultimately, these would enable more equitable and generalizable use of NLP systems for the betterment of healthcare and society.

Corneal neuromediator profiles following laser refractive surgery.

Laser refractive surgery is one of the most commonly performed procedures worldwide. In laser refractive surgery, Femtosecond Laser in Situ Keratomileusis and Refractive Lenticule Extraction have emerged as promising alternatives to microkeratome Laser in Situ Keratomileusis and Photorefractive Keratectomy. Following laser refractive surgery, the corneal nerves, epithelial and stromal cells release neuromediators, including neurotrophins, neuropeptides and neurotransmitters. Notably, nerve growth factor, substance P, calcitonin gene-related peptide and various cytokines are important mediators of neurogenic inflammation and corneal nerve regeneration. Alterations in neuromediator profiles and ocular surface parameters following laser refractive surgery are attributed to the surgical techniques and the severity of tissue insult induced. In this review, we will discuss the (1) Functions of neuromediators and their physiological and clinical significance; (2) Changes in the neuromediators following various laser refractive surgeries; (3) Correlation between neuromediators, ocular surface health and corneal nerve status; and (4) Future directions, including the use of neuromediators as potential biomarkers for ocular surface health following laser refractive surgery, and as adjuncts to aid in corneal regeneration after laser refractive surgery.

Artificial intelligence for anterior segment diseases: Emerging applications in ophthalmology.

With the advancement of computational power, refinement of learning algorithms and architectures, and availability of big data, artificial intelligence (AI) technology, particularly with machine learning and deep learning, is paving the way for 'intelligent' healthcare systems. AI-related research in ophthalmology previously focused on the screening and diagnosis of posterior segment diseases, particularly diabetic retinopathy, age-related macular degeneration and glaucoma. There is now emerging evidence demonstrating the application of AI to the diagnosis and management of a variety of anterior segment conditions. In this review, we provide an overview of AI applications to the anterior segment addressing keratoconus, infectious keratitis, refractive surgery, corneal transplant, adult and paediatric cataracts, angle-closure glaucoma and iris tumour, and highlight important clinical considerations for adoption of AI technologies, potential integration with telemedicine and future directions.

Results of a colorectal enhanced recovery after surgery (ERAS) programme and a qualitative analysis of healthcare workers' perspectives.

BACKGROUND: Enhanced recovery after surgery (ERAS) is a structured programme using a multimodal, evidence-based approach to improve post-operative outcomes. Successful implementation of ERAS can be challenging. We aimed to evaluate our initial experience with colorectal ERAS and explore the perspectives of specialist doctors and nurses. METHODS: From 1 June 2017 to 31 December 2017, all patients who underwent elective colorectal resection and met the ERAS inclusion criteria at the Department of Colorectal Surgery, Singapore General Hospital, were included in the study. Short-term outcomes were compared between patients with >70% compliance to key ERAS components versus those with ≤70% compliance. Department staff were surveyed via questionnaire in July 2019. RESULTS: Three hundred and fifteen patients were included in study. >70% ERAS compliance rate was achieved in 84 patients (26.7%). A higher compliance rate resulted in a significantly shorter length of stay of 6 (IQR 5-8) days vs. 7 (IQR 6-9.5) days (p = 0.025) and lower readmission rate of 3.6% (n = 3) vs. 4.8% (n = 11) (p = 0.042), as well as a trend towards reduced complication rate of 15.4% (n = 13) vs. 22.0% (n = 51) and earlier return to gastrointestinal function. There was a 100% questionnaire response amongst all 12 colorectal surgeons and 5 colorectal resident nurse practitioners. CONCLUSION: Increased adherence to the components of ERAS results in better early outcomes and may have long-term benefits on survival. Effective communication and professional support for the ERAS multi-disciplinary team, as well as understanding healthcare workers' concerns and addressing long-standing practices, is essential for successful implementation of the programme.

Cross-sectional Study on Corneal Denervation in Contralateral Eyes Following SMILE Versus LASIK.

PURPOSE: To compare long-term corneal nerve status following small incision lenticule extraction (SMILE) versus laser in situ keratomileusis (LASIK). METHODS: Twenty-four patients were randomized to receive SMILE in one eye and LASIK in the other eye. In vivo confocal microscopy examination and dry eye assessments were performed at a mean of 4.1 years postoperatively. The patients were further divided into two groups based on the mean assessment time: 2.7 years postoperatively (2.7 years group) and 5.5 years postoperatively (5.5 years group). Another 6 age-matched normal patients were recruited. RESULTS: At 4.1 years, LASIK eyes had significantly less corneal nerve fiber density (CNFD), corneal nerve branch density (CNBD), corneal nerve fiber length (CNFL), and corneal total branch density and significantly more nerves with beading than SMILE eyes. The CNFD, CNBD, CNFL, and number of nerves with sprouting were significantly higher in the 5.5 years group than in the 2.7 years group, in both types of surgery, suggesting persistent nerve regeneration. The CNBD and CNFD in the 5.5 years group, regardless of surgical types, were significantly lower than those in the control group, indicating the nerve status had not recovered to normal ranges even at 5.5 years. High myopic treatment resulted in significantly reduced CNFD with LASIK but not with SMILE. There were no significant differences in the dry eye parameters between the two procedures at 4.1 years postoperatively. CONCLUSIONS: The impact on corneal nerves following refractive surgery is long-lasting. SMILE had better nerve preservation and regeneration than LASIK, but neither procedure had recovered nerve status to normal levels even at 5.5 years. [J Refract Surg. 2020;36(10):653-660.].

Validation of the Use of Automated and Manual Quantitative Analysis of Corneal Nerve Plexus Following Refractive Surgery.

Following refractive surgery, the cornea is denervated and re-innervated, hence a reproducible tool to objectively quantify this change is warranted. This study aimed to determine the repeatability and reproducibility of corneal nerve quantification between automated (ACCMetrics) and manual software (CCMetrics) following refractive surgery. A total of 1007 in vivo confocal microscopy images from 20 post-small incision lenticule extraction (SMILE) or post-laser-assisted in situ keratomileusis (LASIK) patients were evaluated by two independent observers using CCMetrics for corneal nerve fibre density (CNFD), corneal nerve branch density (CNBD), and corneal nerve fibre length (CNFL). Intra-observer and inter-observer reproducibility and repeatability, as well as agreement and correlation between the measurements obtained by ACCMetrics and CCMetrics, were assessed. We found that CNFL demonstrated the best intra- and inter-observer agreement followed by CNFD (intra-class correlation coefficient (ICC) = 0.799 and 0.740, respectively for CNFL; 0.757 and 0.728 for CNFD). CNBD demonstrated poorest intra- and inter-observer ICC. There was an underestimation in ACCMetrics measurements compared to CCMetrics measurements, although the differences were not significant. Our data suggested that both automated and manual methods can be used as reliable tools for the evaluation of corneal nerve status following refractive surgery. However, the measurements obtained with different methods are not interchangeable.