In Nihon Hoshasen Gijutsu Gakkai zasshi

PURPOSE : In this study, we propose a system that combines a depth camera with a deep learning model for estimating the human skeleton and a depth camera to estimate the shooting part to be radiographed and to acquire the thickness of the subject, thereby providing optimized X-ray imaging conditions.

METHODS : We propose a system that provides optimized X-ray imaging conditions by estimating the shooting part and measuring the thickness of the subject using an RGB camera and a depth camera. The system uses OpenPose, a posture estimation library, to estimate the shooting part.

RESULTS : The recognition rate of the shooting part was 15.38% for the depth camera and 84.62% for the RGB camera at a distance of 100 cm, and 42.31% for the depth camera and 100% for the RGB camera at a distance of 120 cm. The measurement accuracy of the subject thickness was within ±10 mm except for a few cases, indicating that the X-ray imaging conditions were optimized for the subject thickness.

CONCLUSION : The implementation of this system in an X-ray system is expected to enable automatic setting of X-ray imaging conditions. The system is also useful in preventing increased exposure dose due to excessive dose or decreased image quality due to insufficient dose caused by incorrect setting of X-ray imaging conditions.

Eto Michihiro, Nakawatari Tomofumi, Hatanaka Yuji

2023-Mar-23

X-ray image, depth camera, dose management, exposure dose, pose estimation

In Journal of the American Board of Family Medicine : JABFM

BACKGROUND : Artificial intelligence (AI) implementation in primary care is limited. Those set to be most impacted by AI technology in this setting should guide it's application. We organized a national deliberative dialogue with primary care stakeholders from across Canada to explore how they thought AI should be applied in primary care.

METHODS : We conducted 12 virtual deliberative dialogues with participants from 8 Canadian provinces to identify shared priorities for applying AI in primary care. Dialogue data were thematically analyzed using interpretive description approaches.

RESULTS : Participants thought that AI should first be applied to documentation, practice operations, and triage tasks, in hopes of improving efficiency while maintaining person-centered delivery, relationships, and access. They viewed complex AI-driven clinical decision support and proactive care tools as impactful but recognized potential risks. Appropriate training and implementation support were the most important external enablers of safe, effective, and patient-centered use of AI in primary care settings.

INTERPRETATION : Our findings offer an agenda for the future application of AI in primary care grounded in the shared values of patients and providers. We propose that, from conception, AI developers work with primary care stakeholders as codesign partners, developing tools that respond to shared priorities.

Upshaw Tara L, Craig-Neil Amy, Macklin Jillian, Gray Carolyn Steele, Chan Timothy C Y, Gibson Jennifer, Pinto Andrew D

2023-Mar-22

Artificial Intelligence, Canada, Family Medicine, Qualitative Research

In Journal of the American Board of Family Medicine : JABFM

PURPOSE : To understand staff and health care providers' views on potential use of artificial intelligence (AI)-driven tools to help care for patients within a primary care setting.

METHODS : We conducted a qualitative descriptive study using individual semistructured interviews. As part of province-wide Learning Health Organization, Community Health Centres (CHCs) are a community-governed, team-based delivery model providing primary care for people who experience marginalization in Ontario, Canada. CHC health care providers and staff were invited to participate. Interviews were audio-recorded and transcribed verbatim. We performed a thematic analysis using a team approach.

RESULTS : We interviewed 27 participants across 6 CHCs. Participants lacked in-depth knowledge about AI. Trust was essential to acceptance of AI; people need to be receptive to using AI and feel confident that the information is accurate. We identified internal influences of AI acceptance, including ease of use and complementing clinical judgment rather than replacing it. External influences included privacy, liability, and financial considerations. Participants felt AI could improve patient care and help prevent burnout for providers; however, there were concerns about the impact on the patient-provider relationship.

CONCLUSIONS : The information gained in this study can be used for future research, development, and integration of AI technology.

Nash Danielle M, Thorpe Cathy, Brown Judith Belle, Kueper Jacqueline K, Rayner Jennifer, Lizotte Daniel J, Terry Amanda L, Zwarenstein Merrick

2023-Mar-22

Artificial Intelligence, Canada, Family Medicine, Informatics, Qualitative Research

In Transplant immunology ; h5-index 20.0

INTRODUCTION : The pathogenesis and progression mechanism of Immunoglobulin A nephropathy (IgAN) is not fully understood. There is a lack of panoramic analysis of IgAN immune cell infiltration and algorithms that are more efficient and accurate for screening key pathogenic genes.

METHODS : RNA sequencing (RNA-seq) data sets on IgAN were downloaded from the Gene Expression Omnibus (GEO) database, including GSE93798, GSE35489, and GSE115857. The RNA-seq data set of kidney tissue as control samples were downloaded from the Genotype-Tissue Expression (GTEx) database. Three machine learning algorithms-weighted gene co-expression network analysis (WGCNA), least absolute shrinkage and selection operator (LASSO), and support vector machine-were used to identify the key pathogenic gene sets of the IgAN disease. The ssGSEA method was applied to calculate the immune cell infiltration (ICI) of IgAN samples, whereas the Spearman test was used for correlation analysis. The receiver operator characteristic curve (ROC) was used to evaluate the diagnostic efficacy of key genes. The correlation between the key genes and ICI was analyzed using the Spearman test.

RESULTS : A total of 177 genes were screened out as differentially expressed genes (DEGs) for IgAN, including 135 up-regulated genes and 42 down-regulated genes. The DEGs were significantly enriched in the inflammatory- or immune-related pathways (gene sets). Activating transcription factor 3 (AFT3), C-X-C Motif Chemokine Ligand 6 (CXCL6), and v-fos FBJ murine osteosarcoma viral oncogene homolog B (FOSB) were identified using WGCNA, support vector machine, and LASSO algorithms. These three genes revealed good diagnostic efficacy in the training and test cohorts. The CXCL6 expression positively correlated with activated B cells and memory B cells.

CONCLUSION : ATF3, FOSB, and CXCL6 genes were identified as potential biomarkers of IgAN. These three genes exhibited good diagnostic efficacy for IgAN. We described the landscape of immune cell infiltration for IgAN. Activated B cells and memory B cells were more highly expressed in the IgAN samples than in the control samples. CXCL6 seems crucial to the pathogenesis of IgAN and may induce IgAN by enriching immune cells. Our study may contribute to developing CXCL6 inhibitors that target B cells for IgAN therapy.

Chen Suzhi, Li Yongzhang, Wang Guangjian, Song Lei, Tan Jinchuan, Yang Fengwen

2023-Mar-20

ATF3, CXLC6, FOSB, Immunoglobulin a nephropathy, Machine learning algorithm

In Acta biomaterialia ; h5-index 89.0

Ding Yuan, Qian Xiaohui, Lin Fenghao, Gao Bingqiang, Wang Weili, Yang Huang, Du Yang, Wang Weilin

2023-Mar-20

NIR-II photothermal therapy, antitumor immunity, biomimetic nanoplatform, immunogenic cell death, tumor-associated macrophages

In Clinica chimica acta; international journal of clinical chemistry

OBJECTIVES : Preoperative identification of malignant adrenal tumors is challenging. 24-h urinary steroid profiling by LC-MS/MS and machine learning has demonstrated high diagnostic power, but the unavailability of bioinformatic models for public use has limited its routine application. We here aimed to increase usability with a novel classification model for the differentiation of adrenocortical adenoma(ACA) and adrenocortical carcinoma(ACC).

METHODS : Eleven steroids (5-pregnenetriol, dehydroepiandrosterone, cortisone, cortisol, α-cortolone, tetrahydro-11-deoxycortisol, etiocholanolone, pregnenolone, pregnanetriol, pregnanediol, and 5-pregnenediol) were quantified by LC-MS/MS in 24-h urine samples from 352 patients with adrenal tumor (281 ACA,71 ACC). Random forest modelling and decision tree algorithms were applied in training (n=188) and test sets (n=80) and independently validated in 84 patients with paired 24-h and spot urine.

RESULTS : After examining different models, a decision tree using excretions of only 5-pregnenetriol and tetrahydro-11-deoxycortisol classified three groups with low, intermediate, and high risk for malignancy. 148/217 ACA were classified as being at low, 67 intermediate, and 2 high risk of malignancy. Conversely, none of the ACC demonstrated a low-risk profile leading to a negative predictive value of 100% for malignancy. In the independent validation cohort, the negative predictive value was again 100% in both 24-h urine and spot urine with a positive predictive value of 87.5% and 86.7%, respectively.

CONCLUSIONS : This simplified LC-MS/MS-based classification model using 24-h-urine provided excellent results for exclusion of ACC and can help to avoid unnecessary surgeries. Analysis of spot urine led to similarly satisfactory results suggesting that cumbersome 24-h urine collection might be dispensable after future validation.

Vogg Nora, Müller Tobias, Floren Andreas, Dandekar Thomas, Riester Anna, Dischinger Ulrich, Kurlbaum Max, Kroiss Matthias, Fassnacht Martin

2023-Mar-20

LC-MS/MS, adrenal tumors, adrenocortical carcinoma, mass spectrometry, steroid profiling