An artificial intelligence approach to classify pathogenic fungal genera of fungal keratitis using corneal confocal microscopy images

Tang, Ningning; Huang, Guangyi; Lei, Daizai; Jiang, Li; Chen, Qi; He, Weiqun; Tang, Fen; Hong, Yiyi; Lv, Jian; Qin, Yuanjun; Lin, Yunru; Lan, Qianqian; Qin, Yikun; Lan, Rushi; Pan, Xipeng; Li, Min; Xu, Fan; Lü, Peng

doi:10.1007/s10792-022-02616-8

Cited by 8 publications

(6 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We therefore support the suggestion that future randomized controlled trials of treatment outcomes should be analysed according to mould genus. If it were possible to distinguish Aspergillus spp from Fusarium spp at the point of first patient contact, using deep learning from images from IVCM [ 53 ] or DNA sequencing, modifying therapy early by the early addition or substitution of voriconazole for natamycin would be an option.…”

Section: Discussionmentioning

confidence: 99%

Antifungal susceptibility profiles for fungal isolates from corneas and contact lenses in the United Kingdom

Tuft,

Stone,

Burton

et al. 2023

Eye

View full text Add to dashboard Cite

Objective To report the identification and results of susceptibility testing for fungal isolates from the cornea or contact lens care systems. Materials and methods In this retrospective epidemiological study, we searched the results of fungal cultures from cornea or contact lens systems referred for identification and susceptibility testing to the United Kingdom National Mycology Reference Laboratory between October 2016 and March 2022. For each fungal isolate, we recorded the genus and species of the fungus and the minimum inhibitory concentration (MIC) to six antifungal agents available to treat corneal infection (amphotericin, econazole, itraconazole, natamycin, posaconazole, and voriconazole). Results There were 600 isolates from 585 patients, comprising 374 (62%) from corneal samples and 226 from contact lenses and care systems, of which 414 (69%) isolates were moulds (filamentous fungi) and 186 (31%) were yeasts. The most frequent moulds isolated were Fusarium spp (234 isolates, 39%) and Aspergillus spp (62, 10%). The most frequent yeasts isolated were Candida spp (112, 19%), predominantly Candida parapsilosis (65, 11%) and Candida albicans (33, 6%), with 35 isolates (6%) of Meyerozyma guilliermondii. In vitro susceptibility was greatest for natamycin (347 moulds tested, mode 4 mg/L, range 0.25–64 mg/L; 98 yeasts tested, mode 4 mg/L, range 0.5–32 mg/L), with susceptibility for 94% for moulds and 99% yeasts. Of the 16 isolates interpreted as highly resistant to natamycin (MIC ≥16 mg/L), 13 were Aspergillus flavus complex. Conclusions In vitro susceptibility supports the use of natamycin for the empiric treatment of fungal keratitis in the UK.

show abstract

Section: Discussionmentioning

confidence: 99%

Antifungal susceptibility profiles for fungal isolates from corneas and contact lenses in the United Kingdom

Tuft,

Stone,

Burton

et al. 2023

Eye

View full text Add to dashboard Cite

show abstract

“…Adopting an MIC of ≥ 16 mg/L as representing an isolate extremely unlikely to respond to natamycin, 16 of 445 (3.6%) of cases in this series would have received an ineffective treatment for approximately the rst 14 days of management until susceptibility results were available. If it were possible to distinguish Aspergillus spp from Fusarium spp at the point of rst patient contact, using deep learning from images from IVCM, 53 or DNA sequencing in the future, modifying therapy by the early addition or substitution of voriconazole for natamycin would be an option. Finally, this data provides a benchmark for monitoring future trends in antifungal resistance in the UK.…”

Section: Discussionmentioning

confidence: 99%

Antifungal susceptibility profiles for fungal isolates from corneas and contact lenses in the United Kingdom

Tuft

Stone²,

Burton

et al. 2023

Preprint

View full text Add to dashboard Cite

Objective To report the identification and results of susceptibility testing for fungal isolates from the cornea or contact lens care systems.Materials and methods In this retrospective epidemiological study, we searched the results of fungal cultures from cornea or contact lens systems referred for identification and susceptibility testing to the United Kingdom National Mycology Reference Laboratory between October 2016 and March 2022. For each fungal isolate, we recorded the genus and species of the fungus and the minimum inhibitory concentration (MIC) to six antifungal agents available to treat corneal infection (amphotericin, econazole, itraconazole, natamycin, posaconazole, and voriconazole).Results There were 600 isolates from 585 patients, comprising 374 (62%) from corneal samples and 226 from contact lenses and care systems, of which 414 (69%) isolates were moulds (filamentous fungi) and 186 (31%) were yeasts. The most frequent moulds isolated were Fusarium spp (234 isolates, 39%) and Aspergillus spp (62, 10%). The most frequent yeasts isolated were Candida spp (112, 19%), predominantly Candida parapsilosis (11%) and Candida albicans (6%), with 35 isolates (6%) of Meyerozyma guilliermondii. In vitro susceptibility was greatest for natamycin (347 moulds tested, mode 4 mg/L, range 0.25-64 mg/L; 98 yeasts tested, mode 4 mg/L, range 0.5–32 mg/L), with susceptibility for 94% for moulds and 99% yeasts. The isolates interpreted as highly resistant to natamycin (N, 14) were predominantly Aspergillus flavus complex.Conclusions In vitro susceptibility supports the use of natamycin for the empiric treatment of fungal keratitis in the UK.

show abstract

“…Certain attempts were made not only to detect fungal hyphae but also to recognize their species. Ningning Tang et al [69] designed an automated method to distinguish Fusarium and Aspergillus genres. To cope with the overfitting phenomenon, they used transfer learning to improve their model's generalization ability.…”

Section: Viral Keratitismentioning

confidence: 99%

Artificial-Intelligence-Enhanced Analysis of In Vivo Confocal Microscopy in Corneal Diseases: A Review

Kryszan,

Wylęgała,

Kijonka

et al. 2024

Diagnostics

View full text Add to dashboard Cite

Artificial intelligence (AI) has seen significant progress in medical diagnostics, particularly in image and video analysis. This review focuses on the application of AI in analyzing in vivo confocal microscopy (IVCM) images for corneal diseases. The cornea, as an exposed and delicate part of the body, necessitates the precise diagnoses of various conditions. Convolutional neural networks (CNNs), a key component of deep learning, are a powerful tool for image data analysis. This review highlights AI applications in diagnosing keratitis, dry eye disease, and diabetic corneal neuropathy. It discusses the potential of AI in detecting infectious agents, analyzing corneal nerve morphology, and identifying the subtle changes in nerve fiber characteristics in diabetic corneal neuropathy. However, challenges still remain, including limited datasets, overfitting, low-quality images, and unrepresentative training datasets. This review explores augmentation techniques and the importance of feature engineering to address these challenges. Despite the progress made, challenges are still present, such as the “black-box” nature of AI models and the need for explainable AI (XAI). Expanding datasets, fostering collaborative efforts, and developing user-friendly AI tools are crucial for enhancing the acceptance and integration of AI into clinical practice.

show abstract

An artificial intelligence approach to classify pathogenic fungal genera of fungal keratitis using corneal confocal microscopy images

Cited by 8 publications

References 33 publications

Antifungal susceptibility profiles for fungal isolates from corneas and contact lenses in the United Kingdom

Antifungal susceptibility profiles for fungal isolates from corneas and contact lenses in the United Kingdom

Antifungal susceptibility profiles for fungal isolates from corneas and contact lenses in the United Kingdom

Artificial-Intelligence-Enhanced Analysis of In Vivo Confocal Microscopy in Corneal Diseases: A Review

Contact Info

Product

Resources

About