Pre-trained convolutional neural networks as feature extractors toward improved malaria parasite detection in thin blood smear images

Rajaraman, Sivaramakrishnan; Antani, Sameer; Poostchi, Mahdieh; Silamut, Kamolrat; Hossain, Md. Amir; Maude, Richard J.; Jaeger, Stefan; Thoma, George R.

doi:10.7717/peerj.4568

Cited by 393 publications

(332 citation statements)

References 38 publications

Supporting

Mentioning

265

Contrasting

Unclassified

Order By: Relevance

“…As an application domain we focus on biological data which come in a huge variety of image types -from fine grained microscopic images to holistic images of plants and animals. Our two case studies focused on applications from the medical and biological field, namely the detection of malaria parasites in thin blood smear images [13] and the detection of stress in tobacco plants used for pharmaceutical purposes [17].…”

Section: Introductionmentioning

confidence: 99%

Effect of Superpixel Aggregation on Explanations in LIME – A Case Study with Biological Data

Schallner

Rabold

Scholz

et al. 2020

Communications in Computer and Information Science

View full text Add to dashboard Cite

End-to-end learning with deep neural networks, such as convolutional neural networks (CNNs), has been demonstrated to be very successful for different tasks of image classification. To make decisions of black-box approaches transparent, different solutions have been proposed. LIME is an approach to explainable AI relying on segmenting images into superpixels based on the Quick-Shift algorithm. In this paper, we present an explorative study of how different superpixel methods, namely Felzenszwalb, SLIC and Compact-Watershed, impact the generated visual explanations. We compare the resulting relevance areas with the image parts marked by a human reference. Results show that image parts selected as relevant strongly vary depending on the applied method. Quick-Shift resulted in the least and Compact-Watershed in the highest correspondence with the reference relevance areas.

show abstract

Section: Introductionmentioning

confidence: 99%

Effect of Superpixel Aggregation on Explanations in LIME – A Case Study with Biological Data

Schallner

Rabold

Scholz

et al. 2020

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…One solution to this are toy models, which can be used to capture complex processes in a simplified and mathematically-tractable model [68]. Results of the comparison between the DeepLabv3 and OpenDevoCell models suggest the need for a specialized pre-trained model [69] optimized for the shape, movement patterns, and intra-cellular contours of a Bacillaria colony (see Figure 2). Specialized pre-trained models have been created for a host of specific types of systems such as linguistic and object recognition and transfer, so creating a model specialized for the analysis of dynamic biological systems is both desirable and attainable.…”

Section: Discussionmentioning

confidence: 99%

Towards a digital diatom: image processing and deep learning analysis ofBacillaria paradoxadynamic morphology

Alicea¹,

Gordon

Harbich³

et al. 2019

Preprint

View full text Add to dashboard Cite

Recent years have witnessed a convergence of data and methods that allow us to approximate the shape, size, and functional attributes of biological organisms. This is not only limited to traditional model species: given the ability to culture and visualize a specific organism, we can capture both its structural and functional attributes. We present a quantitative model for the colonial diatom Bacillaria paradoxa, an organism that presents a number of unique attributes in terms of form and function. To acquire a digital model of B. paradoxa, we extract a series of quantitative parameters from microscopy videos from both primary and secondary sources. These data are then analyzed using a variety of techniques, including two rival deep learning approaches. We provide an overview of neural networks for non-specialists as well as present a series of analysis on Bacillaria phenotype data. The application of deep learning networks allow for two analytical purposes. Application of the DeepLabv3 pre-trained model extracts phenotypic parameters describing the shape of cells constituting Bacillaria colonies. Application of a semantic model trained on nematode embryogenesis data (OpenDevoCell) provides a means to analyze masked images of potential intracellular features. We also advance the analysis of Bacillaria colony movement dynamics by using templating techniques and biomechanical analysis to better understand the movement of individual cells relative to an entire colony. The broader implications of these results are presented, with an eye towards future applications to both hypothesis-driven studies and theoretical advancements in understanding the dynamic morphology of Bacillaria.

show abstract

“…In 2018, Rajaraman et. al [173] used specialized CNN architectures like ResNet for detecting malarial parasites in thin blood smear images. Kang et al [174] improved the performance of 2D CNN by using a 3D multi-view CNN for lung nodule classification using spatial contextual information with the help of 3D Inception-ResNet architecture.…”

Section: Medical Image Processingmentioning

confidence: 99%

A review of deep learning with special emphasis on architectures, applications and recent trends

Sengupta

Basak

Saikia

et al. 2020

Knowledge-Based Systems

303

143

View full text Add to dashboard Cite

Deep learning has solved a problem that as little as five years ago was thought by many to be intractable -the automatic recognition of patterns in data; and it can do so with an accuracy that often surpasses that of human beings. It has solved problems beyond the realm of traditional, hand-crafted machine learning algorithms and captured the imagination of practitioners trying to make sense out of the flood of data that now inundates our society. As public awareness of the efficacy of deep learning increases so does the desire to make use of it. But even for highly trained professionals it can be daunting to approach the rapidly increasing body of knowledge produced by experts in the field. Where does one start? How does one determine if a particular Deep Learning model is applicable to their problem? How does one train and deploy such a network? A primer on the subject can be a good place to start. With that in mind, we present an overview of some of the key multilayer artificial neural networks that comprise deep learning. We also discuss some new automatic architecture optimization protocols that use multi-agent approaches. Further, since guaranteeing system uptime is becoming critical to many computer applications, we include a section on using neural networks for fault detection and subsequent mitigation. This is followed by an exploratory survey of several application areas where deep learning has emerged as a game-changing technology: anomalous behavior detection in financial applications or in financial time-series forecasting, predictive and prescriptive analytics, medical image processing and analysis and power systems research. The thrust of this review is to outline emerging areas of application-oriented research within the deep learning community as well as to provide a handy reference to researchers seeking to use deep learning in their work for what it does best: statistical pattern recognition with unparalleled learning capacity with the ability to scale with information.

show abstract

Pre-trained convolutional neural networks as feature extractors toward improved malaria parasite detection in thin blood smear images

Cited by 393 publications

References 38 publications

Effect of Superpixel Aggregation on Explanations in LIME – A Case Study with Biological Data

Effect of Superpixel Aggregation on Explanations in LIME – A Case Study with Biological Data

Towards a digital diatom: image processing and deep learning analysis ofBacillaria paradoxadynamic morphology

A review of deep learning with special emphasis on architectures, applications and recent trends

Contact Info

Product

Resources

About