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

General General

Low-cost portable microwave sensor for non-invasive monitoring of blood glucose level: novel design utilizing a four-cell CSRR hexagonal configuration.

In Scientific reports ; h5-index 158.0

This article presents a novel design of portable planar microwave sensor for fast, accurate, and non-invasive monitoring of the blood glucose level as an effective technique for diabetes control and prevention. The proposed sensor design incorporates four cells of hexagonal-shaped complementary split ring resonators (CSRRs), arranged in a honey-cell configuration, and fabricated on a thin sheet of an FR4 dielectric substrate.The CSRR sensing elements are coupled via a planar microstrip-line to a radar board operating in the ISM band 2.4-2.5 GHz. The integrated sensor shows an impressive detection capability and a remarkable sensitivity of blood glucose levels (BGLs). The superior detection capability is attributed to the enhanced design of the CSRR sensing elements that expose the glucose samples to an intense interaction with the electromagnetic fields highly concentrated around the sensing region at the induced resonances. This feature enables the developed sensor to detect extremely delicate variations in the electromagnetic properties that characterize the varying-level glucose samples. The desired performance of the fabricated sensor is practically validated through in-vitro measurements using a convenient setup of Vector Network Analyzer (VNA) that records notable traces of frequency-shift responses when the sensor is loaded with samples of 70-120 mg/dL glucose concentrations. This is also demonstrated in the radar-driven prototype where the raw data collected at the radar receiving channel shows obvious patterns that reflect glucose-level variations. Furthermore, the differences in the sensor responses for tested glucose samples are quantified by applying the Principal Component Analysis (PCA) machine learning algorithm. The proposed sensor, beside its impressive detection capability of the diabetes-spectrum glucose levels, has several other favorable attributes including compact size, simple fabrication, affordable cost, non-ionizing nature, and minimum health risk or impact. Such attractive features promote the proposed sensor as a possible candidate for non-invasive glucose levels monitoring for diabetes as evidenced by the preliminary results from a proof-of-concept in-vivo experiment of tracking an individual's BGL by placing his fingertip onto the sensor. The presented system is a developmental platform towards radar-driven wearable continuous BGL monitors.

Omer Ala Eldin, Shaker George, Safavi-Naeini Safieddin, Kokabi Hamid, Alquié Georges, Deshours Frédérique, Shubair Raed M


General General

Multifactorial 10-Year Prior Diagnosis Prediction Model of Dementia.

In International journal of environmental research and public health ; h5-index 73.0

Dementia is a neurodegenerative disorder that affects the older adult population. To date, no cure or treatment to change its course is available. Since changes in the brains of affected individuals could be evidenced as early as 10 years before the onset of symptoms, prognosis research should consider this time frame. This study investigates a broad decision tree multifactorial approach for the prediction of dementia, considering 75 variables regarding demographic, social, lifestyle, medical history, biochemical tests, physical examination, psychological assessment and health instruments. Previous work on dementia prognoses with machine learning did not consider a broad range of factors in a large time frame. The proposed approach investigated predictive factors for dementia and possible prognostic subgroups. This study used data from the ongoing multipurpose Swedish National Study on Aging and Care, consisting of 726 subjects (91 presented dementia diagnosis in 10 years). The proposed approach achieved an AUC of 0.745 and Recall of 0.722 for the 10-year prognosis of dementia. Most of the variables selected by the tree are related to modifiable risk factors; physical strength was important across all ages. Also, there was a lack of variables related to health instruments routinely used for the dementia diagnosis.

Dallora Ana Luiza, Minku Leandro, Mendes Emilia, Rennemark Mikael, Anderberg Peter, Sanmartin Berglund Johan


cost sensitive learning, decision tree, dementia, machine learning, modifiable risk factors, prognosis, wrapper feature selection

Radiology Radiology

Development of a volumetric pancreas segmentation CT dataset for AI applications through trained technologists: a study during the COVID 19 containment phase.

In Abdominal radiology (New York)

PURPOSE : To evaluate the performance of trained technologists vis-à-vis radiologists for volumetric pancreas segmentation and to assess the impact of supplementary training on their performance.

METHODS : In this IRB-approved study, 22 technologists were trained in pancreas segmentation on portal venous phase CT through radiologist-led interactive videoconferencing sessions based on an image-rich curriculum. Technologists segmented pancreas in 188 CTs using freehand tools on custom image-viewing software. Subsequent supplementary training included multimedia videos focused on common errors, which were followed by second batch of 159 segmentations. Two radiologists reviewed all cases and corrected inaccurate segmentations. Technologists' segmentations were compared against radiologists' segmentations using Dice-Sorenson coefficient (DSC), Jaccard coefficient (JC), and Bland-Altman analysis.

RESULTS : Corrections were made in 71 (38%) cases from first batch [26 (37%) oversegmentations and 45 (63%) undersegmentations] and in 77 (48%) cases from second batch [12 (16%) oversegmentations and 65 (84%) undersegmentations]. DSC, JC, false positive (FP), and false negative (FN) [mean (SD)] in first versus second batches were 0.63 (0.15) versus 0.63 (0.16), 0.48 (0.15) versus 0.48 (0.15), 0.29 (0.21) versus 0.21 (0.10), and 0.36 (0.20) versus 0.43 (0.19), respectively. Differences were not significant (p > 0.05). However, range of mean pancreatic volume difference reduced in the second batch [- 2.74 cc (min - 92.96 cc, max 87.47 cc) versus - 23.57 cc (min - 77.32, max 30.19)].

CONCLUSION : Trained technologists could perform volumetric pancreas segmentation with reasonable accuracy despite its complexity. Supplementary training further reduced range of volume difference in segmentations. Investment into training technologists could augment and accelerate development of body imaging datasets for AI applications.

Suman Garima, Panda Ananya, Korfiatis Panagiotis, Edwards Marie E, Garg Sushil, Blezek Daniel J, Chari Suresh T, Goenka Ajit H


Artificial intelligence, COVID-19, Data curation, Deep learning

Radiology Radiology

SCREENet: A Multi-view Deep Convolutional Neural Network for Classification of High-resolution Synthetic Mammographic Screening Scans

ArXiv Preprint

Purpose: To develop and evaluate the accuracy of a multi-view deep learning approach to the analysis of high-resolution synthetic mammograms from digital breast tomosynthesis screening cases, and to assess the effect on accuracy of image resolution and training set size. Materials and Methods: In a retrospective study, 21,264 screening digital breast tomosynthesis (DBT) exams obtained at our institution were collected along with associated radiology reports. The 2D synthetic mammographic images from these exams, with varying resolutions and data set sizes, were used to train a multi-view deep convolutional neural network (MV-CNN) to classify screening images into BI-RADS classes (BI-RADS 0, 1 and 2) before evaluation on a held-out set of exams. Results: Area under the receiver operating characteristic curve (AUC) for BI-RADS 0 vs non-BI-RADS 0 class was 0.912 for the MV-CNN trained on the full dataset. The model obtained accuracy of 84.8%, recall of 95.9% and precision of 95.0%. This AUC value decreased when the same model was trained with 50% and 25% of images (AUC = 0.877, P=0.010 and 0.834, P=0.009 respectively). Also, the performance dropped when the same model was trained using images that were under-sampled by 1/2 and 1/4 (AUC = 0.870, P=0.011 and 0.813, P=0.009 respectively). Conclusion: This deep learning model classified high-resolution synthetic mammography scans into normal vs needing further workup using tens of thousands of high-resolution images. Smaller training data sets and lower resolution images both caused significant decrease in performance.

Saeed Seyyedi, Margaret J. Wong, Debra M. Ikeda, Curtis P. Langlotz


Surgery Surgery

Improvement of nerve imaging speed with coherent anti-Stokes Raman scattering rigid endoscope using deep-learning noise reduction.

In Scientific reports ; h5-index 158.0

A coherent anti-Stokes Raman scattering (CARS) rigid endoscope was developed to visualize peripheral nerves without labeling for nerve-sparing endoscopic surgery. The developed CARS endoscope had a problem with low imaging speed, i.e. low imaging rate. In this study, we demonstrate that noise reduction with deep learning boosts the nerve imaging speed with CARS endoscopy. We employ fine-tuning and ensemble learning and compare deep learning models with three different architectures. In the fine-tuning strategy, deep learning models are pre-trained with CARS microscopy nerve images and retrained with CARS endoscopy nerve images to compensate for the small dataset of CARS endoscopy images. We propose using the equivalent imaging rate (EIR) as a new evaluation metric for quantitatively and directly assessing the imaging rate improvement by deep learning models. The highest EIR of the deep learning model was 7.0 images/min, which was 5 times higher than that of the raw endoscopic image of 1.4 images/min. We believe that the improvement of the nerve imaging speed will open up the possibility of reducing postoperative dysfunction by intraoperative nerve identification.

Yamato Naoki, Niioka Hirohiko, Miyake Jun, Hashimoto Mamoru


General General

Improved haplotype inference by exploiting long-range linking and allelic imbalance in RNA-seq datasets.

In Nature communications ; h5-index 260.0

Haplotype reconstruction of distant genetic variants remains an unsolved problem due to the short-read length of common sequencing data. Here, we introduce HapTree-X, a probabilistic framework that utilizes latent long-range information to reconstruct unspecified haplotypes in diploid and polyploid organisms. It introduces the observation that differential allele-specific expression can link genetic variants from the same physical chromosome, thus even enabling using reads that cover only individual variants. We demonstrate HapTree-X's feasibility on in-house sequenced Genome in a Bottle RNA-seq and various whole exome, genome, and 10X Genomics datasets. HapTree-X produces more complete phases (up to 25%), even in clinically important genes, and phases more variants than other methods while maintaining similar or higher accuracy and being up to 10×  faster than other tools. The advantage of HapTree-X's ability to use multiple lines of evidence, as well as to phase polyploid genomes in a single integrative framework, substantially grows as the amount of diverse data increases.

Berger Emily, Yorukoglu Deniz, Zhang Lillian, Nyquist Sarah K, Shalek Alex K, Kellis Manolis, Numanagić Ibrahim, Berger Bonnie