Research Article: Accurate Detection of Dysmorphic Nuclei Using Dynamic Programming and Supervised Classification

Date Published: January 26, 2017

Publisher: Public Library of Science

Author(s): Marlies Verschuuren, Jonas De Vylder, Hannes Catrysse, Joke Robijns, Wilfried Philips, Winnok H. De Vos, Thomas Abraham.


A vast array of pathologies is typified by the presence of nuclei with an abnormal morphology. Dysmorphic nuclear phenotypes feature dramatic size changes or foldings, but also entail much subtler deviations such as nuclear protrusions called blebs. Due to their unpredictable size, shape and intensity, dysmorphic nuclei are often not accurately detected in standard image analysis routines. To enable accurate detection of dysmorphic nuclei in confocal and widefield fluorescence microscopy images, we have developed an automated segmentation algorithm, called Blebbed Nuclei Detector (BleND), which relies on two-pass thresholding for initial nuclear contour detection, and an optimal path finding algorithm, based on dynamic programming, for refining these contours. Using a robust error metric, we show that our method matches manual segmentation in terms of precision and outperforms state-of-the-art nuclear segmentation methods. Its high performance allowed for building and integrating a robust classifier that recognizes dysmorphic nuclei with an accuracy above 95%. The combined segmentation-classification routine is bound to facilitate nucleus-based diagnostics and enable real-time recognition of dysmorphic nuclei in intelligent microscopy workflows.

Partial Text

Nuclear shape changes are present in a broad range of pathologies. Depending on the origin and cell type, nuclei of cancer cells display strikingly different sizes and overt shape alterations such as grooves, folds or lobes, as compared to normal cells [1,2]. Numerous disorders also demonstrate subtler morphological aberrations such as invaginations or protrusions. These protrusions are often referred to as nuclear blebs and they are characteristic for diseases of the nuclear lamina, i.e., laminopathies [3,4]. In various laminopathies, these blebs represent weak spots, which can sometimes rupture causing illegitimate exchange of nuclear and cytoplasmic proteins [5–8]. Bleb formation has also been observed in viral infections, where it is considered to represent a correlate of nuclear entry and/or egress [9,10]. Despite a clear correlation with disease, not all nuclei in a cell culture display crevices or blebs, and since their formation is time-dependent, it is imperative that they can be automatically detected with high fidelity, preferably in a large number of cells.

Dysmorphic nuclei are characteristic for a wide range of pathologies such as cancer, viral infections and nuclear envelopathies. Automated recognition and analysis of these nuclei may enhance the efficiency of cell-based microscopy experiments aimed at unraveling mechanisms underlying pathology. To this end, we wrote an algorithm that is tailored towards segmentation of dysmorphic nuclei and can be used for a wide variety of cell types acquired with different image modalities. Based on an integrated error score, we have shown that BleND attained a precision that matched the ground truth, when taking into account an inter-individual variability of 7%. The algorithm was further used to build a classifier that accurately predicts whether a nucleus is normal or dysmorphic.




0 0 vote
Article Rating
Notify of
Inline Feedbacks
View all comments