Deep Transfer Learning for Improved Detection of Keratoconus using Corneal Topographic Maps

Ali H. Al-Timemy, Nebras H. Ghaeb, Zahraa M. Musa, Javier Escudero

Research output: Contribution to journalArticlepeer-review


Background/Introduction: Clinical Keratoconus (KCN) detection is a challenging and time-consuming task. In the diagnosis process, ophthalmologist must revise demographic and clinical ophthalmic examinations. The latter include slit-lamb, corneal topographic maps and Pentacam indices (PI).

Methods: We propose an Ensemble of Deep Transfer Learning (EDTL) based on corneal topographic maps. We consider four pretrained networks, SqueezeNet (SqN), AlexNet (AN), ShuffleNet (SfN) and MobileNet-v2 (MN), and fine-tuned them on a dataset of KCN and normal cases, each including four topographic maps. We also consider a PI classifier. Then, our EDTL method combines the output probabilities of each of the five classifiers to obtain a decision based on the fusion of probabilities.

Results: Individually, the classifier based on PI achieved 93.1% accuracy, whereas the deep classifiers reached classification accuracies over 90% only in isolated cases. Overall, the average accuracy of the deep networks over the four corneal maps ranged from 86% (SfN) to 89.9% (AN). The classifier ensemble increased the accuracy of the deep classifiers based on corneal maps to values ranging (92.2% to 93.1%) for SqN and (93.1% to 94.8%) for AN. Including in the ensemble specific combinations of corneal maps’ classifiers and PI increased the accuracy to 98.3%. Moreover, visualization of first learner filters in the networks and Grad-CAMs confirmed that the networks had learned relevant clinical features.

Conclusions: This study shows the potential of creating ensembles of deep classifiers fine-tuned with a transfer learning strategy as it resulted in an improved accuracy while showing learnable filters and Grad-CAMs that agree with clinical knowledge. This is a step further towards the potential clinical deployment of an improved computerassisted diagnosis system for KCN detection to help ophthalmologists to confirm the clinical decision and to
perform fast and accurate KCN treatment.
Original languageEnglish
Number of pages16
JournalCognitive Computation
Early online date16 Jun 2021
Publication statusE-pub ahead of print - 16 Jun 2021


  • AlexNet
  • Computer Aided Diagnosis (CAD)
  • Corneal topographic maps
  • Deep learning
  • Keratoconus
  • Probability fusion
  • Transfer learning


Dive into the research topics of 'Deep Transfer Learning for Improved Detection of Keratoconus using Corneal Topographic Maps'. Together they form a unique fingerprint.

Cite this