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

Radiology Radiology

Noninvasive KRAS mutation estimation in colorectal cancer using a deep learning method based on CT imaging.

In BMC medical imaging

BACKGROUND : The detection of Kirsten rat sarcoma viral oncogene homolog (KRAS) gene mutations in colorectal cancer (CRC) is key to the optimal design of individualized therapeutic strategies. The noninvasive prediction of the KRAS status in CRC is challenging. Deep learning (DL) in medical imaging has shown its high performance in diagnosis, classification, and prediction in recent years. In this paper, we investigated predictive performance by using a DL method with a residual neural network (ResNet) to estimate the KRAS mutation status in CRC patients based on pre-treatment contrast-enhanced CT imaging.

METHODS : We have collected a dataset consisting of 157 patients with pathology-confirmed CRC who were divided into a training cohort (n = 117) and a testing cohort (n = 40). We developed an ResNet model that used portal venous phase CT images to estimate KRAS mutations in the axial, coronal, and sagittal directions of the training cohort and evaluated the model in the testing cohort. Several groups of expended region of interest (ROI) patches were generated for the ResNet model, to explore whether tissues around the tumor can contribute to cancer assessment. We also explored a radiomics model with the random forest classifier (RFC) to predict KRAS mutations and compared it with the DL model.

RESULTS : The ResNet model in the axial direction achieved the higher area under the curve (AUC) value (0.90) in the testing cohort and peaked at 0.93 with an input of 'ROI and 20-pixel' surrounding area. AUC of radiomics model in testing cohorts were 0.818. In comparison, the ResNet model showed better predictive ability.

CONCLUSIONS : Our experiments reveal that the computerized assessment of the pre-treatment CT images of CRC patients using a DL model has the potential to precisely predict KRAS mutations. This new model has the potential to assist in noninvasive KRAS mutation estimation.

He Kan, Liu Xiaoming, Li Mingyang, Li Xueyan, Yang Hualin, Zhang Huimao


Colorectal Neoplasm, Deep learning, Mutation

General General

CRISPRpred(SEQ): a sequence-based method for sgRNA on target activity prediction using traditional machine learning.

In BMC bioinformatics

BACKGROUND : The latest works on CRISPR genome editing tools mainly employs deep learning techniques. However, deep learning models lack explainability and they are harder to reproduce. We were motivated to build an accurate genome editing tool using sequence-based features and traditional machine learning that can compete with deep learning models.

RESULTS : In this paper, we present CRISPRpred(SEQ), a method for sgRNA on-target activity prediction that leverages only traditional machine learning techniques and hand-crafted features extracted from sgRNA sequences. We compare the results of CRISPRpred(SEQ) with that of DeepCRISPR, the current state-of-the-art, which uses a deep learning pipeline. Despite using only traditional machine learning methods, we have been able to beat DeepCRISPR for the three out of four cell lines in the benchmark dataset convincingly (2.174%, 6.905% and 8.119% improvement for the three cell lines).

CONCLUSION : CRISPRpred(SEQ) has been able to convincingly beat DeepCRISPR in 3 out of 4 cell lines. We believe that by exploring further, one can design better features only using the sgRNA sequences and can come up with a better method leveraging only traditional machine learning algorithms that can fully beat the deep learning models.

Muhammad Rafid Ali Haisam, Toufikuzzaman Md, Rahman Mohammad Saifur, Rahman M Sohel


CRISPR, Cas9, Deep learning, Machine learning, sgRNA

General General

Smartphone-Based Self-Testing of COVID-19 Using Breathing Sounds.

In Telemedicine journal and e-health : the official journal of the American Telemedicine Association

Telemedicine could be a key to control the world-wide disruptive and spreading novel coronavirus disease (COVID-19) pandemic. The COVID-19 virus directly targets the lungs, leading to pneumonia-like symptoms and shortness of breath with life-threatening consequences. Despite the fact that self-quarantine and social distancing are indispensable during the pandemic, the procedure for testing COVID-19 contraction is conventionally available through nasal swabs, saliva test kits, and blood work at healthcare settings. Therefore, devising personalized self-testing kits for COVID-19 virus and other similar viruses is heavily admired. Many e-health initiatives have been made possible by the advent of smartphones with embedded software, hardware, high-performance computing, and connectivity capabilities. A careful review of breathing sounds and their implications in identifying breathing complications suggests that the breathing sounds of COVID-19 contracted users may reveal certain acoustic signal patterns, which is worth investigating. To this end, acquiring respiratory data solely from breathing sounds fed to the smartphone's microphone strikes as a very appealing resolution. The acquired breathing sounds can be analyzed using advanced signal processing and analysis in tandem with new deep/machine learning and pattern recognition techniques to separate the breathing phases, estimate the lung volume, oxygenation, and to further classify the breathing data input into healthy or unhealthy cases. The ideas presented have the potential to be deployed as self-test breathing monitoring apps for the ongoing global COVID-19 pandemic, where users can check their breathing sound pattern frequently through the app.

Faezipour Miad, Abuzneid Abdelshakour


e-health, home health monitoring, sensor technology, technology, telemedicine

General General

Single-Trial EEG Responses Classified Using Latency Features.

In International journal of neural systems

Covert attention has been repeatedly shown to impact on EEG responses after single and repeated practice sessions. Machine learning techniques are increasingly adopted to classify single-trial EEG responses thereby primarily relying on amplitude-based features instead of latency-based features. In this study, we investigated changes in EEG response signatures of nine healthy older subjects when performing 10 sessions of covert attention training. We show that, when we trained classifiers to distinguish recorded EEG patterns between the two experimental conditions (a target stimulus is "present" or "not present"), latency-based classifiers outperform the amplitude-based ones and that classification accuracy improved along with behavioral accuracy, providing supportive evidence of brain plasticity.

Hardiansyah Irzam, Pergher Valentina, Van Hulle Marc M


EEG, Longitudinal covert attention training, latency features, machine learning classification

General General

EHR-Independent Predictive Decision Support Architecture Based on OMOP.

In Applied clinical informatics ; h5-index 22.0

BACKGROUND :  The increasing availability of molecular and clinical data of cancer patients combined with novel machine learning techniques has the potential to enhance clinical decision support, example, for assessing a patient's relapse risk. While these prediction models often produce promising results, a deployment in clinical settings is rarely pursued.

OBJECTIVES :  In this study, we demonstrate how prediction tools can be integrated generically into a clinical setting and provide an exemplary use case for predicting relapse risk in melanoma patients.

METHODS :  To make the decision support architecture independent of the electronic health record (EHR) and transferable to different hospital environments, it was based on the widely used Observational Medical Outcomes Partnership (OMOP) common data model (CDM) rather than on a proprietary EHR data structure. The usability of our exemplary implementation was evaluated by means of conducting user interviews including the thinking-aloud protocol and the system usability scale (SUS) questionnaire.

RESULTS :  An extract-transform-load process was developed to extract relevant clinical and molecular data from their original sources and map them to OMOP. Further, the OMOP WebAPI was adapted to retrieve all data for a single patient and transfer them into the decision support Web application for enabling physicians to easily consult the prediction service including monitoring of transferred data. The evaluation of the application resulted in a SUS score of 86.7.

CONCLUSION :  This work proposes an EHR-independent means of integrating prediction models for deployment in clinical settings, utilizing the OMOP CDM. The usability evaluation revealed that the application is generally suitable for routine use while also illustrating small aspects for improvement.

Unberath Philipp, Prokosch Hans Ulrich, Gr√ľndner Julian, Erpenbeck Marcel, Maier Christian, Christoph Jan


General General

Optimizing neuromodulation based on surrogate neural states for seizure suppression in a rat temporal lobe epilepsy model.

In Journal of neural engineering ; h5-index 52.0

OBJECTIVE : Developing a new neuromodulation method for epilepsy treatment requires a large amount of time and resources to find effective stimulation parameters and often fails due to inter-subject variability in stimulation effect. As an alternative, we present a novel data-driven surrogate approach which can optimize the neuromodulation efficiently by investigating the stimulation effect on surrogate neural states.

APPROACH : Medial septum (MS) optogenetic stimulation was applied for modulating electrophysiological activities of the hippocampus in a rat temporal lobe epilepsy model. For the new approach, we implemented machine learning techniques to describe the pathological neural states and to optimize the stimulation parameters. Specifically, first, we found neural state surrogates to estimate a seizure susceptibility based on hippocampal local field potentials. Second, we modulated the neural state surrogates in a desired way with the subject-specific optimal stimulation parameters found by in vivo Bayesian optimization. Finally, we tested whether modulating the neural state surrogates affected seizure frequency.

MAIN RESULTS : We found two neural state surrogates: The first was hippocampal theta power by considering its well-known relationship with epilepsy, and the second was the output of pre-ictal state model (PriSM) which was built by characterizing the hippocampal activity during pre-ictal period. The optimal stimulation parameters found by Bayesian optimization outperformed the other parameters in terms of modulating the surrogates toward anti-seizure neural state. When treatment efficacy was tested, the subject-specific optimal parameters for increasing theta power were more effective to suppress seizures than fixed stimulation parameter (7Hz). However, modulation of the other neural state surrogate, PriSM, did not suppress seizures.

SIGNIFICANCE : The surrogate approach can save enormous time and resources to find subject-specific optimal stimulation parameters which can effectively modulate neural states and further improve therapeutic effectiveness. This approach also can be used for improving neuromodulation treatment of other neurological or psychiatric diseases.

Park Sang-Eon, Connolly Mark J, Exarchos Ioannis, Fernandez Alejandra, Ghetiya Mihir, Gutekunst Claire-Anne, Gross Robert E


data-driven, epilepsy, medial septum, optimization, optogenetics, surrogate