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

Public Health Public Health

An interpretable machine learning framework for measuring urban perceptions from panoramic street view images.

In iScience

The proliferation of street view images (SVIs) and the constant advancements in deep learning techniques have enabled urban analysts to extract and evaluate urban perceptions from large-scale urban streetscapes. However, many existing analytical frameworks have been found to lack interpretability due to their end-to-end structure and "black-box" nature, thereby limiting their value as a planning support tool. In this context, we propose a five-step machine learning framework for extracting neighborhood-level urban perceptions from panoramic SVIs, specifically emphasizing feature and result interpretability. By utilizing the MIT Place Pulse data, the developed framework can systematically extract six dimensions of urban perceptions from the given panoramas, including perceptions of wealth, boredom, depression, beauty, safety, and liveliness. The practical utility of this framework is demonstrated through its deployment in Inner London, where it was used to visualize urban perceptions at the Output Area (OA) level and to verify against real-world crime rate.

Liu Yunzhe, Chen Meixu, Wang Meihui, Huang Jing, Thomas Fisher, Rahimi Kazem, Mamouei Mohammad

2023-Mar-17

Artificial Intelligence, Environmental sciences

General General

Conv-CapsNet: capsule based network for COVID-19 detection through X-Ray scans.

In Multimedia tools and applications

Coronavirus, a virus that spread worldwide rapidly and was eventually declared a pandemic. The rapid spread made it essential to detect Coronavirus infected people to control the further spread. Recent studies show that radiological images such as X-Rays and CT scans provide essential information in detecting infection using deep learning models. This paper proposes a shallow architecture based on Capsule Networks with convolutional layers to detect COVID-19 infected persons. The proposed method combines the ability of the capsule network to understand spatial information with convolutional layers for efficient feature extraction. Due to the model's shallow architecture, it has 23M parameters to train and requires fewer training samples. The proposed system is fast and robust and correctly classifies the X-Ray images into three classes, i.e. COVID-19, No Findings, and Viral Pneumonia. Experimental results on the X-Ray dataset show that our model performs well despite having fewer samples for the training and achieved an average accuracy of 96.47% for multi-class and 97.69% for binary classification on 5-fold cross-validation. The proposed model would be useful to researchers and medical professionals for assistance and prognosis for COVID-19 infected patients.

Sharma Pulkit, Arya Rhythm, Verma Richa, Verma Bindu

2023-Feb-21

COVID-19 detection, Capsule networks, Chest X-ray classification, Medical imaging

Public Health Public Health

Noninvasive Detection of Stress by Biochemical Profiles from the Skin.

In ACS sensors

Stress is a leading cause of several disease types, yet it is underdiagnosed as current diagnostic methods are mainly based on self-reporting and interviews that are highly subjective, inaccurate, and unsuitable for monitoring. Although some physiological measurements exist (e.g., heart rate variability and cortisol), there are no reliable biological tests that quantify the amount of stress and monitor it in real time. In this article, we report a novel way to measure stress quickly, noninvasively, and accurately. The overall detection approach is based on measuring volatile organic compounds (VOCs) emitted from the skin in response to stress. Sprague Dawley male rats (n = 16) were exposed to underwater trauma. Sixteen naive rats served as a control group (n = 16). VOCs were measured before, during, and after induction of the traumatic event, by gas chromatography linked with mass spectrometry determination and quantification, and an artificially intelligent nanoarray for easy, inexpensive, and portable sensing of the VOCs. An elevated plus maze during and after the induction of stress was used to evaluate the stress response of the rats, and machine learning was used for the development and validation of a computational stress model at each time point. A logistic model classifier with stepwise selection yielded a 66-88% accuracy in detecting stress with a single VOC (2-hydroxy-2-methyl-propanoic acid), and an SVM (support vector machine) model showed a 66-72% accuracy in detecting stress with the artificially intelligent nanoarray. The current study highlights the potential of VOCs as a noninvasive, automatic, and real-time stress predictor for mental health.

Mansour Elias, Palzur Eilam, Broza Yoav Y, Saliba Walaa, Kaisari Sharon, Goldstein Pavel, Shamir Alon, Haick Hossam

2023-Feb-27

behavioral, sensor, skin, stress, volatile organic compound

oncology Oncology

Prognostic Significance of Tumor-Infiltrating Lymphocytes Determined Using LinkNet on Colorectal Cancer Pathology Images.

In JCO precision oncology

PURPOSE : Tumor-infiltrating lymphocytes (TILs) have a significant prognostic value in cancers. However, very few automated, deep learning-based TIL scoring algorithms have been developed for colorectal cancer (CRC).

MATERIALS AND METHODS : We developed an automated, multiscale LinkNet workflow for quantifying TILs at the cellular level in CRC tumors using H&E-stained images from the Lizard data set with annotations of lymphocytes. The predictive performance of the automatic TIL scores (TILsLink) for disease progression and overall survival (OS) was evaluated using two international data sets, including 554 patients with CRC from The Cancer Genome Atlas (TCGA) and 1,130 patients with CRC from Molecular and Cellular Oncology (MCO).

RESULTS : The LinkNet model provided outstanding precision (0.9508), recall (0.9185), and overall F1 score (0.9347). Clear continuous TIL-hazard relationships were observed between TILsLink and the risk of disease progression or death in both TCGA and MCO cohorts. Both univariate and multivariate Cox regression analyses for the TCGA data demonstrated that patients with high TIL abundance had a significant (approximately 75%) reduction in risk for disease progression. In both the MCO and TCGA cohorts, the TIL-high group was significantly associated with improved OS in univariate analysis (30% and 54% reduction in risk, respectively). The favorable effects of high TIL levels were consistently observed in different subgroups (classified according to known risk factors).

CONCLUSION : The proposed deep-learning workflow for automatic TIL quantification on the basis of LinkNet can be a useful tool for CRC. TILsLink is likely an independent risk factor for disease progression and carries predictive information of disease progression beyond the current clinical risk factors and biomarkers. The prognostic significance of TILsLink for OS is also evident.

Liu Anran, Li Xingyu, Wu Hongyi, Guo Bangwei, Jonnagaddala Jitendra, Zhang Hong, Xu Steven

2023-Feb

General General

Raw Electroencephalogram-Based Cognitive Workload Classification Using Directed and Nondirected Functional Connectivity Analysis and Deep Learning.

In Big data

With the phenomenal rise in internet-of-things devices, the use of electroencephalogram (EEG) based brain-computer interfaces (BCIs) can empower individuals to control equipment with thoughts. These allow BCI to be used and pave the way for pro-active health management and the development of internet-of-medical-things architecture. However, EEG-based BCIs have low fidelity, high variance, and EEG signals are very noisy. These challenges compel researchers to design algorithms that can process big data in real-time while being robust to temporal variations and other variations in the data. Another issue in designing a passive BCI is the regular change in user's cognitive state (measured through cognitive workload). Though considerable amount of research has been conducted on this front, methods that could withstand high variability in EEG data and still reflect the neuronal dynamics of cognitive state variations are lacking and much needed in literature. In this research, we evaluate the efficacy of a combination of functional connectivity algorithms and state-of-the-art deep learning algorithms for the classification of three different levels of cognitive workload. We acquire 64-channel EEG data from 23 participants executing the n-back task at three different levels; 1-back (low-workload condition), 2-back (medium-workload condition), and 3-back (high-workload condition). We compared two different functional connectivity algorithms, namely phase transfer entropy (PTE) and mutual information (MI). PTE is a directed functional connectivity algorithm, whereas MI is non-directed. Both methods are suitable for extracting functional connectivity matrices in real-time, which could eventually be used for rapid, robust, and efficient classification. For classification, we use the recently proposed BrainNetCNN deep learning model, designed specifically to classify functional connectivity matrices. Results reveal a classification accuracy of 92.81% with MI and BrainNetCNN and a staggering 99.50% with PTE and BrainNetCNN on test data. PTE can yield a higher classification accuracy due to its robustness to linear mixing of the data and its ability to detect functional connectivity across a range of analysis lags.

Gupta Anmol, Daniel Ronnie, Rao Akash, Roy Partha Pratim, Chandra Sushil, Kim Byung-Gyu

2023-Feb-27

BrainNetCNN, cognitive workload, functional connectivity analysis, mutual information, passive BCI, phase transfer entropy

General General

Beware of Simple Methods for Structure-Based Virtual Screening: The Critical Importance of Broader Comparisons.

In Journal of chemical information and modeling

We discuss how data unbiasing and simple methods such as protein-ligand Interaction FingerPrint (IFP) can overestimate virtual screening performance. We also show that IFP is strongly outperformed by target-specific machine-learning scoring functions, which were not considered in a recent report concluding that simple methods were better than machine-learning scoring functions at virtual screening.

Tran-Nguyen Viet-Khoa, Ballester Pedro J

2023-Feb-27