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

General General

Is it safe to lift COVID-19 travel bans? The Newfoundland story.

In Computational mechanics

A key strategy to prevent a local outbreak during the COVID-19 pandemic is to restrict incoming travel. Once a region has successfully contained the disease, it becomes critical to decide when and how to reopen the borders. Here we explore the impact of border reopening for the example of Newfoundland and Labrador, a Canadian province that has enjoyed no new cases since late April, 2020. We combine a network epidemiology model with machine learning to infer parameters and predict the COVID-19 dynamics upon partial and total airport reopening, with perfect and imperfect quarantine conditions. Our study suggests that upon full reopening, every other day, a new COVID-19 case would enter the province. Under the current conditions, banning air travel from outside Canada is more efficient in managing the pandemic than fully reopening and quarantining 95% of the incoming population. Our study provides quantitative insights of the efficacy of travel restrictions and can inform political decision making in the controversy of reopening.

Linka Kevin, Rahman Proton, Goriely Alain, Kuhl Ellen


COVID-19, Epidemiology, Machine learning, Reproduction number, SEIR model

General General

COVID CT-Net: Predicting Covid-19 From Chest CT Images Using Attentional Convolutional Network

ArXiv Preprint

The novel corona-virus disease (COVID-19) pandemic has caused a major outbreak in more than 200 countries around the world, leading to a severe impact on the health and life of many people globally. As of Aug 25th of 2020, more than 20 million people are infected, and more than 800,000 death are reported. Computed Tomography (CT) images can be used as a as an alternative to the time-consuming "reverse transcription polymerase chain reaction (RT-PCR)" test, to detect COVID-19. In this work we developed a deep learning framework to predict COVID-19 from CT images. We propose to use an attentional convolution network, which can focus on the infected areas of chest, enabling it to perform a more accurate prediction. We trained our model on a dataset of more than 2000 CT images, and report its performance in terms of various popular metrics, such as sensitivity, specificity, area under the curve, and also precision-recall curve, and achieve very promising results. We also provide a visualization of the attention maps of the model for several test images, and show that our model is attending to the infected regions as intended. In addition to developing a machine learning modeling framework, we also provide the manual annotation of the potentionally infected regions of chest, with the help of a board-certified radiologist, and make that publicly available for other researchers.

Shakib Yazdani, Shervin Minaee, Rahele Kafieh, Narges Saeedizadeh, Milan Sonka


oncology Oncology

Computer-Aided Diagnosis Systems in Diagnosing Malignant Thyroid Nodules on Ultrasonography: A Systematic Review and Meta-Analysis.

In European thyroid journal

Background : Computer-aided diagnosis (CAD) systems are being applied to the ultrasonographic diagnosis of malignant thyroid nodules, but it remains controversial whether the systems add any accuracy for radiologists.

Objective : To determine the accuracy of CAD systems in diagnosing malignant thyroid nodules.

Methods : PubMed, EMBASE, and the Cochrane Library were searched for studies on the diagnostic performance of CAD systems. The diagnostic performance was assessed by pooled sensitivity and specificity, and their accuracy was compared with that of radiologists. The present systematic review was registered in PROSPERO (CRD42019134460).

Results : Nineteen studies with 4,781 thyroid nodules were included. Both the classic machine learning- and the deep learning-based CAD system had good performance in diagnosing malignant thyroid nodules (classic machine learning: sensitivity 0.86 [95% CI 0.79-0.92], specificity 0.85 [95% CI 0.77-0.91], diagnostic odds ratio (DOR) 37.41 [95% CI 24.91-56.20]; deep learning: sensitivity 0.89 [95% CI 0.81-0.93], specificity 0.84 [95% CI 0.75-0.90], DOR 40.87 [95% CI 18.13-92.13]). The diagnostic performance of the deep learning-based CAD system was comparable to that of the radiologists (sensitivity 0.87 [95% CI 0.78-0.93] vs. 0.87 [95% CI 0.85-0.89], specificity 0.85 [95% CI 0.76-0.91] vs. 0.87 [95% CI 0.81-0.91], DOR 40.12 [95% CI 15.58-103.33] vs. DOR 44.88 [95% CI 30.71-65.57]).

Conclusions : The CAD systems demonstrated good performance in diagnosing malignant thyroid nodules. However, experienced radiologists may still have an advantage over CAD systems during real-time diagnosis.

Xu Lei, Gao Junling, Wang Quan, Yin Jichao, Yu Pengfei, Bai Bin, Pei Ruixia, Chen Dingzhang, Yang Guochun, Wang Shiqi, Wan Mingxi


Artificial intelligence, Thyroid nodule, Ultrasonography

General General

Diabetic Kidney Disease: Challenges, Advances, and Opportunities.

In Kidney diseases (Basel, Switzerland)

Background : Diabetic kidney disease (DKD) is the most common cause of the end-stage renal disease (ESRD). Regardless of intensive treatments with hyperglycemic control, blood pressure control, and the use of renin-angiotensin system blockades, the prevalence of DKD remains high. Recent studies suggest that the spectrum of DKD has been changed and many progresses have been made to develop new treatments for DKD. Therefore, it is time to perform a systemic review on the new developments in the field of DKD.

Summary : Although the classic clinical presentation of DKD is characterized by a slow progression from microalbuminuria to macroalbuminuria and by a hyperfiltration at the early stage and progressive decline of renal function at the late stage, recent epidemiological studies suggest that DKD patients have a variety of clinical presentations and progression rates to ESRD. Some DKD patients have a decline in renal function without albuminuria but display prominent vascular and interstitial fibrosis on renal histology. DKD patients are more susceptible to acute kidney injury, which might contribute to the interstitial fibrosis. A large portion of type 2 diabetic patients with albuminuria could have overlapping nondiabetic glomerular disease, and therefore, kidney biopsy is required for differential diagnosis for these patients. Only a small portion of DKD patients eventually progress to end-stage renal failure. However, we do not have sensitive and specific biomarkers to identify these high-risk patients. Genetic factors that have a strong association with DKD progression have not been identified yet. A combination of circulating tumor necrosis factor receptor (TNFR)1, TNFR2, and kidney injury molecular 1 provides predictive value for DKD progression. Artificial intelligence could enhance the predictive values for DKD progression by combining the clinical parameters and biological markers. Sodium-glucose co-transporter-2 inhibitors should be added to the new standard care of DKD patients. Several promising new drugs are in clinical trials.

Key Messages : Over last years, our understanding of DKD has been much improved and new treatments to halt the progression of DKD are coming. However, better diagnostic tools, predictive markers, and treatment options are still urgently needed to help us to better manage these patients with this detrimental disease.

Chen Ya, Lee Kyung, Ni Zhaohui, He John Cijiang


Diabetic kidney disease, Inflammation, Nonproteinuric pathway, Progression, Treatment

Radiology Radiology

Machine Learning Prediction of Radiofrequency Thermal Ablation Efficacy: A New Option to Optimize Thyroid Nodule Selection.

In European thyroid journal

Background : Radiofrequency (RF) is a therapeutic modality for reducing the volume of large benign thyroid nodules. If thermal therapies are interpreted as an alternative strategy to surgery, critical issues in their use are represented by the extent of nodule reduction and by the durability of nodule reduction over a long period of time.

Objective : To assess the ability of machine learning to discriminate nodules with volume reduction rate (VRR) < or ≥50% at 12 months following RF treatment.

Methods : A machine learning model was trained with a dataset of 402 cytologically benign thyroid nodules subjected to RF at six Italian Institutions. The model was trained with the following variables: baseline nodule volume, echostructure, macrocalcalcifications, vascularity, and 12-month VRR.

Results : After training, the model could distinguish between nodules having VRR <50% from those having VRR ≥50% in 85% of cases (accuracy: 0.85; 95% confidence interval [CI]: 0.80-0.90; sensitivity: 0.70; 95% CI: 0.62-0.75; specificity: 0.99; 95% CI: 0.98-1.0; positive predictive value: 0.95; 95% CI: 0.92-0.98; negative predictive value: 0.95; 95% CI: 0.92-0.98).

Conclusions : This study demonstrates that a machine learning model can reliably identify those nodules that will have VRR < or ≥50% at 12 months after one RF treatment session. Predicting which nodules will be poor or good responders represents valuable data that may help physicians and patients decide on the best treatment option between thermal ablation and surgery or in predicting if more than one session might be necessary to obtain a significant volume reduction.

Negro Roberto, Rucco Matteo, Creanza Annalisa, Mormile Alberto, Limone Paolo Piero, Garberoglio Roberto, Spiezia Stefano, Monti Salvatore, Cugini Christian, El Dalati Ghassan, Deandrea Maurilio


Artificial Intelligence, Laser, Machine learning, Nodules, Radiofrequency, Thermal ablation, Thyroid

General General

Hand-Gesture Recognition Based on EMG and Event-Based Camera Sensor Fusion: A Benchmark in Neuromorphic Computing.

In Frontiers in neuroscience ; h5-index 72.0

Hand gestures are a form of non-verbal communication used by individuals in conjunction with speech to communicate. Nowadays, with the increasing use of technology, hand-gesture recognition is considered to be an important aspect of Human-Machine Interaction (HMI), allowing the machine to capture and interpret the user's intent and to respond accordingly. The ability to discriminate between human gestures can help in several applications, such as assisted living, healthcare, neuro-rehabilitation, and sports. Recently, multi-sensor data fusion mechanisms have been investigated to improve discrimination accuracy. In this paper, we present a sensor fusion framework that integrates complementary systems: the electromyography (EMG) signal from muscles and visual information. This multi-sensor approach, while improving accuracy and robustness, introduces the disadvantage of high computational cost, which grows exponentially with the number of sensors and the number of measurements. Furthermore, this huge amount of data to process can affect the classification latency which can be crucial in real-case scenarios, such as prosthetic control. Neuromorphic technologies can be deployed to overcome these limitations since they allow real-time processing in parallel at low power consumption. In this paper, we present a fully neuromorphic sensor fusion approach for hand-gesture recognition comprised of an event-based vision sensor and three different neuromorphic processors. In particular, we used the event-based camera, called DVS, and two neuromorphic platforms, Loihi and ODIN + MorphIC. The EMG signals were recorded using traditional electrodes and then converted into spikes to be fed into the chips. We collected a dataset of five gestures from sign language where visual and electromyography signals are synchronized. We compared a fully neuromorphic approach to a baseline implemented using traditional machine learning approaches on a portable GPU system. According to the chip's constraints, we designed specific spiking neural networks (SNNs) for sensor fusion that showed classification accuracy comparable to the software baseline. These neuromorphic alternatives have increased inference time, between 20 and 40%, with respect to the GPU system but have a significantly smaller energy-delay product (EDP) which makes them between 30× and 600× more efficient. The proposed work represents a new benchmark that moves neuromorphic computing toward a real-world scenario.

Ceolini Enea, Frenkel Charlotte, Shrestha Sumit Bam, Taverni Gemma, Khacef Lyes, Payvand Melika, Donati Elisa


electromyography (EMG) signal processing, event-based camera, hand-gesture classification, neuromorphic engineering, sensor fusion, spiking neural networks (SNNs)