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

Public Health Public Health

Patient Journey Map to Improve the Home Isolation Experience of Persons with Mild COVID-19 Symptoms: Design Research for Service Touchpoints of Artificial Intelligence in eHealth.

In JMIR medical informatics ; h5-index 23.0

BACKGROUND : In the context of COVID-19 outbreak, 80% of the persons are those with mild symptoms who are required to self-recover at home. They have a strong demand for remote healthcare that despite the great potential of artificial intelligence are not met in the current (e)-health services. Understanding the real needs of these persons is lacking.

OBJECTIVE : The aim of this paper is to contribute with a fine grained understanding of the home isolation experience of persons with mild COVID-19 symptoms, in order to enhance AI in eHealth services.

METHODS : Design research in which a qualitative approach was used to map the patient journey. Data on the home isolation experiences of persons with mild COVID-19 symptoms was collected from top viewed personal video stories on YouTube and their additional comment threads. For the analysis this data was transcribed, coded and mapped into the patient journey map.

RESULTS : The key findings on the home isolation experience of persons with mild COVID-19 symptoms concern: (a) Considerable awareness period before testing positive and home-recovery period; (b) Less generic but more personal symptoms experiences; (c) Negative mood experience curve; (d) Inadequate home healthcare service support for mild COVID-19 patients through all stages. (e) Benefits and drawbacks of Social media support for mild COVID-19 patients; (f) Several touchpoint needs for home healthcare interaction with AI.

CONCLUSIONS : The design of the patient journey map and underlying insights on the home isolation experience of persons with mild COVID-19 symptoms serves Health - and IT professionals to more effectively apply AI technology into eHealth services for mild Covid-19 patients, for which three main service concepts are proposed: (I) Trustful public health information to release stress; (II) Personal Covid-19 health monitoring. (III) Community Support.

He Qian, Du Fei, Simonse Lianne W L


Radiology Radiology

Preserving image texture while reducing radiation dose with a deep learning image reconstruction algorithm in chest CT: A phantom study.

In Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB)

PURPOSE : To assess whether a deep learning image reconstruction algorithm (TrueFidelity) can preserve the image texture of conventional filtered back projection (FBP) at reduced dose levels attained by ASIR-V in chest CT.

METHODS : Phantom images were acquired using a clinical chest protocol (7.6 mGy) and two levels of dose reduction (60% and 80%). Images were reconstructed with FBP, ASIR-V (50% and 100% blending) and TrueFidelity (low (DL-L), medium (DL-M) and high (DL-H) strength). Noise (SD), noise power spectrum (NPS) and task-based transfer function (TTF) were calculated. Noise texture was quantitatively compared by computing root-mean-square deviations (RMSD) of NPS with respect to FBP. Four experienced readers performed a contrast-detail evaluation. The dose reducing potential of TrueFidelity compared to ASIR-V was assessed by fitting SD and contrast-detail as a function of dose.

RESULTS : DL-M and DL-H reduced noise and NPS area compared to FBP and 50% ASIR-V, at all dose levels. At 7.6 mGy, NPS of ASIR-V 50/100% was shifted towards lower frequencies (fpeak = 0.22/0.13 mm-1, RMSD = 0.14/0.38), with respect to FBP (fpeak = 0.30 mm-1). Marginal difference was observed for TrueFidelity: fpeak = 0.33/0.30/0.30 mm-1 and RMSD = 0.03/0.04/0.07 for L/M/H strength. Values of TTF50% were independent of DL strength and higher compared to FBP and ASIR-V, at all dose and contrast levels. Contrast-detail was highest for DL-H at all doses. Compared to 50% ASIR-V, DL-H had an estimated dose reducing potential of 50% on average, without impairing noise, texture and detectability.

CONCLUSIONS : TrueFidelity preserves the image texture of FBP, while outperforming ASIR-V in terms of noise, spatial resolution and detectability at lower doses.

Franck Caro, Zhang Guozhi, Deak Paul, Zanca Federica


Chest, Computed tomography, Contrast-detail evaluation, Deep learning image reconstruction, Dosimetry, Image quality, Iterative reconstruction

General General

Individualized identification of first-episode bipolar disorder using machine learning and cognitive tests.

In Journal of affective disorders ; h5-index 79.0

Identifying cognitive dysfunction in the early stages of Bipolar Disorder (BD) can allow for early intervention. Previous studies have shown a strong correlation between cognitive dysfunction and number of manic episodes. The objective of this study was to apply machine learning (ML) techniques on a battery of cognitive tests to identify first-episode BD patients (FE-BD). Two cohorts of participants were used for this study. Cohort #1 included 74 chronic BD patients (CHR-BD) and 53 healthy controls (HC), while the Cohort #2 included 37 FE-BD and 18 age- and sex-matched HC. Cognitive functioning was assessed using the Cambridge Neuropsychological Test Automated Battery (CANTAB). The tests examined domains of visual processing, spatial memory, attention and executive function. We trained an ML model to distinguish between chronic BD patients (CHR-BD) and HC at the individual level. We used linear Support Vector Machines (SVM) and were able to identify individual CHR-BD patients at 77% accuracy. We then applied the model to Cohort #2 (FE-BD patients) and achieved an accuracy of 76% (AUC = 0.77). These results reveal that cognitive impairments may appear in early stages of BD and persist into later stages. This suggests that the same deficits may exist for both CHR-BD and FE-BD. These cognitive deficits may serve as markers for early BD. Our study provides a tool that these early markers can be used for detection of BD.

Sawalha Jeffrey, Cao Liping, Chen Jianshan, Selvitella Alessandro, Liu Yang, Yang Chanjuan, Li Xuan, Zhang Xiaofei, Sun Jiaqi, Zhang Yamin, Zhao Liansheng, Cui Liqian, Zhang Yizhi, Sui Jie, Greiner Russell, Li Xin-Min, Greenshaw Andrew, Li Tao, Cao Bo


General General

Automated multi-class classification of skin lesions through deep convolutional neural network with dermoscopic images.

In Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

As an analytic tool in medicine, deep learning has gained great attention and opened new ways for disease diagnosis. Recent studies validate the effectiveness of deep learning algorithms for binary classification of skin lesions (i.e., melanomas and nevi classes) with dermoscopic images. Nonetheless, those binary classification methods cannot be applied to the general clinical situation of skin cancer screening in which multi-class classification must be taken into account. The main objective of this research is to develop, implement, and calibrate an advanced deep learning model in the context of automated multi-class classification of skin lesions. The proposed Deep Convolutional Neural Network (DCNN) model is carefully designed with several layers, and multiple filter sizes, but fewer filters and parameters to improve efficacy and performance. Dermoscopic images are acquired from the International Skin Imaging Collaboration databases (ISIC-17, ISIC-18, and ISIC-19) for experiments. The experimental results of the proposed DCNN approach are presented in terms of precision, sensitivity, specificity, and other metrics. Specifically, it attains 94 % precision, 93 % sensitivity, and 91 % specificity in ISIC-17. It is demonstrated by the experimental results that this proposed DCNN approach outperforms state-of-the-art algorithms, exhibiting 0.964 area under the receiver operating characteristics (AUROC) in ISIC-17 for the classification of skin lesions and can be used to assist dermatologists in classifying skin lesions. As a result, this proposed approach provides a novel and feasible way for automating and expediting the skin lesion classification task as well as saving effort, time, and human life.

Iqbal Imran, Younus Muhammad, Walayat Khuram, Kakar Mohib Ullah, Ma Jinwen


Artificial intelligence, Computer vision, Convolutional neural network, Deep learning, Dermoscopy, Image processing, Melanomas, Nevi, Pattern recognition, Skin cancer screening, Skin lesion classification

Surgery Surgery

Predicting hearing recovery following treatment of idiopathic sudden sensorineural hearing loss with machine learning models.

In American journal of otolaryngology ; h5-index 23.0

PURPOSE : Idiopathic sudden sensorineural hearing loss (ISSHL) is an emergency otological disease, and its definite prognostic factors remain unclear. This study applied machine learning methods to develop a new ISSHL prognosis prediction model.

MATERIALS AND METHODS : This retrospective study reviewed the medical data of 244 patients who underwent combined intratympanic and systemic steroid treatment for ISSHL at a tertiary referral center between January 2015 and October 2019. We used 35 variables to predict hearing recovery based on Siegel's criteria. In addition to performing an analysis based on the conventional logistic regression model, we developed prediction models with five machine learning methods: least absolute shrinkage and selection operator, decision tree, random forest (RF), support vector machine, and boosting. To compare the predictive ability of each model, the accuracy, precision, recall, F-score, and the area under the receiver operator characteristic curves (ROC-AUC) were calculated.

RESULTS : Former otological history, ear fullness, delay between symptom onset and treatment, delay between symptom onset and intratympanic steroid injection (ITSI), and initial hearing thresholds of the affected and unaffected ears differed significantly between the recovery and non-recovery groups. While the RF method (accuracy: 72.22%, ROC-AUC: 0.7445) achieved the highest predictive power, the other methods also featured relatively good predictive power. In the RF model, the following variables were identified to be important for hearing-recovery prediction: delay between symptom onset and ITSI or the initial treatment, initial hearing levels of the affected and non-affected ears, body mass index, and a previous history of hearing loss.

CONCLUSIONS : The machine learning models predictive of hearing recovery following treatment for ISSHL showed superior predictive power relative to the conventional logistic regression method, potentially allowing for better patient treatment outcomes.

Uhm Taewoong, Lee Jae Eun, Yi Seongbaek, Choi Sung Won, Oh Se Joon, Kong Soo Keun, Lee Il Woo, Lee Hyun Min


Hearing loss, Machine learning, Outcome prediction, Prognosis, Sudden

General General

Prediction of suicide attempts in a prospective cohort study with a nationally representative sample of the US population.

In Psychological medicine ; h5-index 82.0

BACKGROUND : There is still little knowledge of objective suicide risk stratification.

METHODS : This study aims to develop models using machine-learning approaches to predict suicide attempt (1) among survey participants in a nationally representative sample and (2) among participants with lifetime major depressive episodes. We used a cohort called the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) that was conducted in two waves and included a nationally representative sample of the adult population in the United States. Wave 1 involved 43 093 respondents and wave 2 involved 34 653 completed face-to-face reinterviews with wave 1 participants. Predictor variables included clinical, stressful life events, and sociodemographic variables from wave 1; outcome included suicide attempt between wave 1 and wave 2.

RESULTS : The model built with elastic net regularization distinguished individuals who had attempted suicide from those who had not with an area under the ROC curve (AUC) of 0.89, balanced accuracy 81.86%, specificity 89.22%, and sensitivity 74.51% for the general population. For participants with lifetime major depressive episodes, AUC was 0.89, balanced accuracy 81.64%, specificity 85.86%, and sensitivity 77.42%. The most important predictor variables were a diagnosis of borderline personality disorder, post-traumatic stress disorder, and being of Asian descent for the model in all participants; and previous suicide attempt, borderline personality disorder, and overnight stay in hospital because of depressive symptoms for the model in participants with lifetime major depressive episodes. Random forest and artificial neural networks had similar performance.

CONCLUSIONS : Risk for suicide attempt can be estimated with high accuracy.

Machado Cristiane Dos Santos, Ballester Pedro L, Cao Bo, Mwangi Benson, Caldieraro Marco Antonio, Kapczinski Flávio, Passos Ives Cavalcante


NESARC, Suicide, depression, machine learning, prediction