Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Health is the Motive and Digital is the Instrument.

In Journal of the Indian Institute of Science

The coronavirus crisis has seen an unprecedented response from India and the world. If the viral outbreak has exposed gross inadequacies in the healthcare systems of nations both rich and poor, it has stirred a digital healthcare revolution that has been building since the past decade. We have seen how this new era of digital health evolved over the years since healthcare started getting increasingly unaffordable in the western countries forcing a relook in their strategies to explosion of digital innovations in mobile telephony and applications, internet, wearable devices, artificial intelligence, robotics, big data and genomics. The single biggest trigger for the digital shift has indeed been the COVID-19 pandemic this year, more so in India with astonishing response from the private enterprise and the proactive push from the government so evident. However, the full potential of this digital revolution cannot be realized as long as core structural reforms in public healthcare do not take place along with significant boost in digital infrastructure. The way digital technologies have helped facilitate strategy and response to the global pandemic and with predictions of more zoonotic outbreaks impending in the coming years, it has become imperative for the world to increasingly adopt and integrate digital innovations to make healthcare more accessible, interconnected and affordable.

Seethalakshmi S, Nandan Rahul

2020-Sep-27

General General

Deep Representation Learning of Patient Data from Electronic Health Records (EHR): A Systematic Review

ArXiv Preprint

Patient representation learning refers to learning a dense mathematical representation of a patient that encodes meaningful information from Electronic Health Records (EHRs). This is generally performed using advanced deep learning methods. This study presents a systematic review of this field and provides both qualitative and quantitative analyses from a methodological perspective. We identified studies developing patient representations from EHRs with deep learning methods from MEDLINE, EMBASE, Scopus, the Association for Computing Machinery (ACM) Digital Library, and Institute of Electrical and Electronics Engineers (IEEE) Xplore Digital Library. After screening 362 articles, 48 papers were included for a comprehensive data collection. We noticed a typical workflow starting with feeding raw data, applying deep learning models, and ending with clinical outcome predictions as evaluations of the learned representations. Specifically, learning representations from structured EHR data was dominant (36 out of 48 studies). Recurrent Neural Networks were widely applied as the deep learning architecture (LSTM: 13 studies, GRU: 11 studies). Disease prediction was the most common application and evaluation (30 studies). Benchmark datasets were mostly unavailable (28 studies) due to privacy concerns of EHR data, and code availability was assured in 20 studies. We show the importance and feasibility of learning comprehensive representations of patient EHR data through a systematic review. Advances in patient representation learning techniques will be essential for powering patient-level EHR analyses. Future work will still be devoted to leveraging the richness and potential of available EHR data. Knowledge distillation and advanced learning techniques will be exploited to assist the capability of learning patient representation further.

Yuqi Si, Jingcheng Du, Zhao Li, Xiaoqian Jiang, Timothy Miller, Fei Wang, W. Jim Zheng, Kirk Roberts

2020-10-06

General General

Wavelength weightings in machine learning for ovine joint tissue differentiation using diffuse reflectance spectroscopy (DRS).

In Biomedical optics express

Objective: To investigate the DRS of ovine joint tissue to determine the optimal optical wavelengths for tissue differentiation and relate these wavelengths to the biomolecular composition of tissues. In this study, we combine machine learning with DRS for tissue classification and then look further at the weighting matrix of the classifier to further understand the key differentiating features. Methods: Supervised machine learning was used to analyse DRS data. After normalising the data, dimension reduction was achieved through multiclass Fisher's linear discriminant analysis (Multiclass FLDA) and classified with linear discriminant analysis (LDA). The classifier was first run with all the tissue types and the wavelength range 190 nm - 1081 nm. We analysed the weighting matrix of the classifier and then ran the classifier again, the first time using the ten highest weighted wavelengths and the second using only the single highest. Our method was applied to a dataset containing ovine joint tissue including cartilage, cortical and subchondral bone, fat, ligament, meniscus, and muscle. Results: It achieved a classification accuracy of 100% using the wavelength 190 nm - 1081 nm (2048 attributes) with an accuracy of 90% being present for 10 attributes with the exception of those with comparable compositions such as ligament and meniscus. An accuracy greater than 70% was achieved using a single wavelength, with the same exceptions. Conclusion: Multiclass FLDA combined with LDA is a viable technique for tissue identification from DRS data. The majority of differentiating features existed within the wavelength ranges 370-470 and 800-1010 nm. Focusing on key spectral regions means that a spectrometer with a narrower range can potentially be used, with less computational power needed for subsequent analysis.

Gunaratne Rajitha, Goncalves Joshua, Monteath Isaac, Sheh Raymond, Kapfer Michael, Chipper Richard, Robertson Brett, Khan Riaz, Fick Daniel, Ironside Charles N

2020-Sep-01

General General

Machine learning for composition analysis of ssDNA using chemical enhancement in SERS.

In Biomedical optics express

Surface-enhanced Raman spectroscopy (SERS) is an attractive method for bio-chemical sensing due to its potential for single molecule sensitivity and the prospect of DNA composition analysis. In this manuscript we leverage metal specific chemical enhancement effect to detect differences in SERS spectra of 200-base length single-stranded DNA (ssDNA) molecules adsorbed on gold or silver nanorod substrates, and then develop and train a linear regression as well as neural network models to predict the composition of ssDNA. Our results indicate that employing substrates of different metals that host a given adsorbed molecule leads to distinct SERS spectra, allowing to probe metal-molecule interactions under distinct chemical enhancement regimes. Leveraging this difference and combining spectra from different metals as an input for PCA (Principal Component Analysis) and NN (Neural Network) models, allows to significantly lower the detection errors compared to manual feature-choosing analysis as well as compared to the case where data from single metal is used. Furthermore, we show that NN model provides superior performance in the presence of complex noise and data dispersion factors that affect SERS signals collected from metal substrates fabricated on different days.

Nguyen Phuong H L, Hong Brandon, Rubin Shimon, Fainman Yeshaiahu

2020-Sep-01

General General

Real-time retinal layer segmentation of OCT volumes with GPU accelerated inferencing using a compressed, low-latency neural network.

In Biomedical optics express

Segmentation of retinal layers in optical coherence tomography (OCT) is an essential step in OCT image analysis for screening, diagnosis, and assessment of retinal disease progression. Real-time segmentation together with high-speed OCT volume acquisition allows rendering of en face OCT of arbitrary retinal layers, which can be used to increase the yield rate of high-quality scans, provide real-time feedback during image-guided surgeries, and compensate aberrations in adaptive optics (AO) OCT without using wavefront sensors. We demonstrate here unprecedented real-time OCT segmentation of eight retinal layer boundaries achieved by 3 levels of optimization: 1) a modified, low complexity, neural network structure, 2) an innovative scheme of neural network compression with TensorRT, and 3) specialized GPU hardware to accelerate computation. Inferencing with the compressed network U-NetRT took 3.5 ms, improving by 21 times the speed of conventional U-Net inference without reducing the accuracy. The latency of the entire pipeline from data acquisition to inferencing was only 41 ms, enabled by parallelized batch processing. The system and method allow real-time updating of en face OCT and OCTA visualizations of arbitrary retinal layers and plexuses in continuous mode scanning. To the best our knowledge, our work is the first demonstration of an ophthalmic imager with embedded artificial intelligence (AI) providing real-time feedback.

Borkovkina Svetlana, Camino Acner, Janpongsri Worawee, Sarunic Marinko V, Jian Yifan

2020-Jul-01

General General

UNMIX-ME: spectral and lifetime fluorescence unmixing via deep learning.

In Biomedical optics express

Hyperspectral fluorescence lifetime imaging allows for the simultaneous acquisition of spectrally resolved temporal fluorescence emission decays. In turn, the acquired rich multidimensional data set enables simultaneous imaging of multiple fluorescent species for a comprehensive molecular assessment of biotissues. However, to enable quantitative imaging, inherent spectral overlap between the considered fluorescent probes and potential bleed-through must be considered. Such a task is performed via either spectral or lifetime unmixing, typically independently. Herein, we present "UNMIX-ME" (unmix multiple emissions), a deep learning-based fluorescence unmixing routine, capable of quantitative fluorophore unmixing by simultaneously using both spectral and temporal signatures. UNMIX-ME was trained and validated using an in silico framework replicating the data acquisition process of a compressive hyperspectral fluorescent lifetime imaging platform (HMFLI). It was benchmarked against a conventional LSQ method for tri and quadri-exponential simulated samples. Last, UNMIX-ME's potential was assessed for NIR FRET in vitro and in vivo preclinical applications.

Smith Jason T, Ochoa Marien, Intes Xavier

2020-Jul-01