MYCOLab Ai Logo - gold and army green

Evaluation of the DendrisKIT®DP for the Diagnosis of Superficial Fungal Infections

Summary

Researchers evaluated a new rapid test called DendrisKIT®DP that can identify skin, nail, and hair fungal infections much faster than traditional methods. The test uses PCR technology combined with artificial intelligence to detect 13 different fungal species directly from patient samples in less than 48 hours, compared to traditional cultures that take up to 4 weeks. The test showed good accuracy with 84% sensitivity and 89% specificity, and when combined with traditional methods, provides a better overall diagnostic strategy for fungal infections.

Background

Superficial fungal infections affecting skin, hair, and nails are frequently diagnosed using conventional microscopic examination and culture, which have limitations including low sensitivity and prolonged turnaround times of up to four weeks. Over the past two decades, various PCR techniques have been developed to overcome these limitations, but many do not allow complete species identification or have limited detection capabilities.

Objective

To evaluate the DendrisKIT®DP, a new molecular diagnostic test combining pan-fungal PCR, DNA microchip technology, and machine learning algorithms for rapid diagnosis of superficial fungal infections directly from clinical samples.

Results

Among 57 samples with dermatophytes and/or Candida albicans culture, DendrisKIT®DP was positive for 44 samples (77.2%), negative for 10 samples (17.5%), and had invalid PCR for 3 samples. Overall sensitivity was 83.9% (95% CI: 72.8-91.0%) and specificity was 88.9% (95% CI: 67.2-96.9%) with concordance in 65 samples (76.5%).

Conclusion

The DendrisKIT®DP demonstrates satisfactory diagnostic performance for superficial fungal infections with faster results than conventional culture methods. A diagnostic algorithm combining DendrisKIT®DP with conventional methods for negative results is recommended to ensure optimal sensitivity, and the test’s performance is likely to improve with continued algorithm refinement.
  • Published in:Journal of Fungi,
  • Study Type:Retrospective Evaluation Study,
  • Source: PMID: 40278090
Scroll to Top