Article
Medicine, General & Internal
Lina Ren, Junxian Liang, Feng Wan, Yongjun Wang, Xi-Jian Dai
Summary: A practical risk score tool was developed to predict individual dementia risk, providing guidance for individuals to identify their potential risk profile and take precise actions for dementia delay or prevention.
Article
Health Care Sciences & Services
Juyoung Shin, Joonyub Lee, Taehoon Ko, Kanghyuck Lee, Yera Choi, Hun-Sung Kim
Summary: This study proposes a diabetes prediction model based on machine learning (ML) to encourage individuals at risk of diabetes to employ healthy interventions. The XGBoost model shows the best predictive performance among different algorithms compared.
JOURNAL OF PERSONALIZED MEDICINE
(2022)
Review
Psychiatry
Gonzalo Salazar de Pablo, Erich Studerus, Julio Vaquerizo-Serrano, Jessica Irving, Ana Catalan, Dominic Oliver, Helen Baldwin, Andrea Danese, Seena Fazel, Ewout W. Steyerberg, Daniel Stahl, Paolo Fusar-Poli
Summary: The impact of precision psychiatry on clinical practice has not been systematically evaluated. Validated prediction models are available to support the diagnosis, prognosis, and treatment response prediction of psychiatric conditions, particularly psychosis. However, there is a lack of implementation research in real-world clinical practice.
SCHIZOPHRENIA BULLETIN
(2021)
Review
Immunology
May Y. Choi, Karen H. Costenbader
Summary: There is evidence that patients with systemic lupus erythematosus (SLE) have markers of inflammation and autoimmunity before the diagnosis. Understanding the preclinical phase of SLE can help identify at-risk patients and implement preventative strategies.
FRONTIERS IN IMMUNOLOGY
(2022)
Article
Business
Li Dong, Haichao Zheng, Liting Li, Linna Hao
Summary: This study proposes a framework of human-machine hybrid prediction market and evaluates its effectiveness using economic analytical models and numerical simulations. The study provides four design guidelines, including the impact of machines' collaboration behavior on the prediction market, the advantages of introducing machines as competitors, the effects of co-competition between machines and humans on prediction performance, and the influence of correct trust in machine models on market performance.
ELECTRONIC COMMERCE RESEARCH AND APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Emer Brady, Mathias Wullum Nielsen, Jens Peter Andersen, Sabine Oertelt-Prigione
Summary: The study found that sex-disaggregated analyses are infrequently presented or planned in COVID-19 studies, with only a small percentage including sex as an analytical variable. This highlights the importance of considering sex and gender differences in research designs for COVID-19 studies.
NATURE COMMUNICATIONS
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Vignesh A. Arasu, Laurel A. Habel, Ninah S. Achacoso, Diana S. M. Buist, Jason B. Cord, Laura J. Esserman, Nola M. Hylton, M. Maria Glymour, John Kornak, Lawrence H. Kushi, Donald A. Lewis, Vincent X. Liu, Caitlin M. Lydon, Diana L. Miglioretti, Daniel A. Navarro, Albert Pu, Li Shen, Weiva Sieh, Hyo-Chun Yoon, Catherine Lee
Summary: By comparing selected mammography artificial intelligence algorithms and the Breast Cancer Surveillance Consortium risk model, it was found that AI algorithms performed better in predicting 5-year risk, and combining the two could further improve prediction accuracy.
Article
Multidisciplinary Sciences
David A. Kolin, Scott Kulm, Olivier Elemento
Summary: The study found that both clinical and genetic factors contribute to the risk of venous thromboembolism. By combining clinical factors and genetic variants into a scoring system, the prediction of venous thromboembolism risk is more accurate.
SCIENTIFIC REPORTS
(2021)
Article
Law
Samantha Besson
Summary: The emergence of dual-use technologies and their potential impact on humanity has raised concerns about anticipating both the risks and benefits of scientific progress. The human right to science provides a framework for understanding and addressing these concerns. This special issue aims to address the specific duties and responsibilities derived from the right to anticipate and protect against the adverse effects of science.
INTERNATIONAL JOURNAL OF HUMAN RIGHTS
(2023)
Review
Biochemical Research Methods
Samilla B. Rezende, Lucas R. Lima, Maria L. R. Macedo, Octavio L. Franco, Marlon H. Cardoso
Summary: Peptides and proteins play crucial roles in biological processes. Determining their three-dimensional structures using experimental and computational methods has greatly advanced structure prediction. Machine learning and deep learning approaches have further improved the accuracy of structure prediction. These methods have greatly assisted in answering key biological questions.
CURRENT BIOINFORMATICS
(2023)
Article
Pharmacology & Pharmacy
Zhenxiang Gao, Maria Gorenflo, David C. Kaelber, Vincent M. Monnier, Rong Xu
Summary: Diabetes mellitus (DM) increases the risk of age-related cataracts, and there is currently no medication approved to delay cataract progression. Researchers used AI to search for drugs that could suppress cataract surgery. They combined AI-based candidate drug prediction with clinical data from electronic health records of cataract patients to identify potential drugs. Aspirin, melatonin, ibuprofen, and acetylcysteine were found to be associated with a reduced risk of cataract extraction in different diabetes patient groups. These drugs have the potential to delay cataract progression by inhibiting cyclooxygenase-2 (COX-2).
FRONTIERS IN PHARMACOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Alexander Belikov, Andrey Rzhetsky, James Evans
Summary: The growth of published science has made it difficult for both human and algorithmic agents to reason over prior knowledge and select the next experiment. Uncertainty about the reproducibility of published findings further increases this challenge. The use of massive digital archives and automated experiments allows us to computationally evaluate these challenges and identify opportunities to accelerate scientific progress. By developing a Bayesian calculus, the authors demonstrate the ability to predict robust scientific claims with findings from published literature, weighted by factors that increase replicability. This approach identifies bias and reveals that research activity with scientific focus and social and institutional diversity is more likely to replicate.
NATURE MACHINE INTELLIGENCE
(2022)
Article
Biochemical Research Methods
Akshaya V. Annapragada, Joseph L. Greenstein, Sanjukta N. Bose, Bradford D. Winters, Sridevi V. Sarma, Raimond L. Winslow
Summary: Hypoxemia is a major contributing factor to mortality in critically ill patients, including those with conditions like brain injury, cardiac arrest, and COVID-19. The SWIFT deep learning model can accurately predict blood oxygen saturation waveforms in the future, providing valuable information for clinical interventions and patient management.
PLOS COMPUTATIONAL BIOLOGY
(2021)
Article
Oncology
Deshan Liu, Dixiang Song, Weihai Ning, Yuduo Guo, Ting Lei, Yanming Qu, Mingshan Zhang, Chunyu Gu, Haoran Wang, Junpeng Ji, Yongfei Wang, Yao Zhao, Nidan Qiao, Hongwei Zhang
Summary: This study developed a clinical prediction model for predicting the risk of VTE after neurosurgery using advanced machine learning techniques and statistical methods. The model, which includes 8 clinical factors and biomarkers, showed high accuracy in both internal and external validation.
Article
Oncology
Lauren G. Johnson, Rakiya Saidu, Cecilia Svanholm-Barrie, Rosalind Boa, Jennifer Moodley, Ana Tergas, David Persing, Scott A. Campbell, Wei-Yann Tsai, Thomas C. Wright, Lynette Denny, Louise Kuhn
Summary: This study evaluated the use of cancer biomarkers in cervical cancer screening and found that they can help differentiate between women with and without CIN2(+) when used as reflex tests after a positive HPV result. However, these biomarkers did not outperform an approach utilizing more stringent HPV testing and genotype selection in improving the sensitivity/specificity balance.
CANCER EPIDEMIOLOGY BIOMARKERS & PREVENTION
(2022)
Article
Endocrinology & Metabolism
Antti-Mathias Taka, Taina Harkonen, Paula Vahasalo, Johanna Lempainen, Riitta Veijola, Jorma Ilonen, Mikael Knip
Summary: The study divided participants into two groups based on the T1D risk conferred by their HLA genotype. Group 1 participants were younger at diagnosis and had more family members affected by T1D, while Group 2 had a higher frequency of DKA, longer duration of symptoms before diagnosis, and higher hemoglobin A1c levels at diagnosis. Group 1 showed more autoantibodies, while Group 2 had a higher percentage of autoantibody negativity.
PEDIATRIC DIABETES
(2022)
Article
Endocrinology & Metabolism
Emily K. Sims, Rachel E. J. Besser, Colin Dayan, Cristy Geno Rasmussen, Carla Greenbaum, Kurt J. Griffin, William Hagopian, Mikael Knip, Anna E. Long, Frank Martin, Chantal Mathieu, Marian Rewers, Andrea K. Steck, John M. Wentworth, Stephen S. Rich, Olga Kordonouri, Anette-Gabriele Ziegler, Kevan C. Herold
Summary: Most screening programs for type 1 diabetes identify individuals at risk through relatives, but many patients don't have a family history. Recent advancements in disease-modifying therapies have sparked interest in population screening. Existing programs rely on genetic or autoantibody screening, which have provided valuable insights into disease progression and screening timing.
Article
Endocrinology & Metabolism
Jorma Ilonen, Antti-Pekka Laine, Minna Kiviniemi, Taina Harkonen, Johanna Lempainen, Mikael Knip
Summary: The study aimed to characterize the demographics and genetic associations of type 1 diabetes endotypes defined by the first appearing islet specific autoantibodies. The analysis of children diagnosed with type 1 diabetes before the age of 10 years revealed significant differences in demographic and genetic features between endotypes defined by glutamic acid decarboxylase (GADA) or insulin autoantibodies (IAA) as the first autoantibody. The results support the assumption that the first autoantibody can be deduced based on islet autoantibody combinations, and strong differences were observed in sex, age, and genetic associations between GADA-initiated and IAA-initiated autoimmunity.
PEDIATRIC DIABETES
(2022)
Correction
Multidisciplinary Sciences
Jelena Stsepetova, Kart Simre, Aili Tagoma, Oivi Uibo, Aleksandr Peet, Heli Siljander, Vallo Tillmann, Mikael Knip, Reet Mandar, Raivo Uibo
SCIENTIFIC REPORTS
(2022)
Article
Multidisciplinary Sciences
Jelena Stsepetova, Kart Simre, Aili Tagoma, Oivi Uibo, Aleksandr Peet, Heli Siljander, Vallo Tillmann, Mikael Knip, Reet Mandar, Raivo Uibo
Summary: This study compared the microbiota composition and immune markers concentration in breast milk from mothers of children with genetic predisposition to celiac disease (CD) and mothers of unaffected children. The breast milk of mothers with genetically predisposed offspring who developed CD showed higher bacterial phylotype abundance and diversity, as well as a different bacterial composition. The immune markers were associated with specific bacterial genera and species, and may influence the risk for development of CD beyond early childhood.
SCIENTIFIC REPORTS
(2022)
Editorial Material
Pediatrics
Mikael Knip
Article
Nutrition & Dietetics
Leena Hakola, Maarit Oikarinen, Sari Niinisto, David Cuthbertson, Jussi Lehtonen, Leena Puustinen, Amir-Babak Sioofy-Khojine, Jarno Honkanen, Mikael Knip, Jeffrey P. Krischer, Iris Erlund, Heikki Hyoty, Suvi M. Virtanen
Summary: This study found that serum fatty acid and vitamin D levels in infancy are associated with the risk of microbial infections at 18 months of age. Specifically, higher proportions of n-3 polyunsaturated fatty acids and docosapentaenoic acid were associated with a decreased risk of coxsackievirus B2 and respiratory syncytial virus infections, while vitamin D concentration was not consistently associated with infection risk.
CLINICAL NUTRITION
(2022)
Article
Endocrinology & Metabolism
Olga Kordonouri, David Cuthbertson, Malin Belteky, Barbel Aschemeier-Fuchs, Neil H. White, Elisabeth Cummings, Mikael Knip, Johnny Ludvigsson
Summary: Viral infections early in life may initiate the autoimmune process or later development of type 1 diabetes, while certain bacterial infections appeared to increase the risk of both multiple autoantibodies and clinical type 1 diabetes.
Article
Pediatrics
Vilma Kielevainen, Maaret Turtinen, Kristiina Luopajarvi, Taina Harkonen, Jorma Ilonen, Mikael Knip
Summary: A register-based retrospective cohort study of 4993 Finnish children diagnosed with type 1 diabetes revealed that HLA class II genes are associated with disease manifestation. Higher disease susceptibility is linked to younger age at diagnosis and shorter duration of symptoms before diagnosis.
Article
Endocrinology & Metabolism
Minna Harsunen, Jarno L. T. Kettunen, Taina Harkonen, Om Dwivedi, Mikko Lehtovirta, Paula Vahasalo, Riitta Veijola, Jorma Ilonen, Paivi J. Miettinen, Mikael Knip, Tiinamaija Tuomi
Summary: More than 10% of AAB-negative children with diabetes were found to have monogenic diabetes through genetic testing. Genetic diagnosis can lead to major changes in treatment, therefore, it is recommended to refer all AAB-negative pediatric diabetes patients for genetic testing.
Article
Multidisciplinary Sciences
Astrid Oras, Henna Kallionpa, Tomi Suomi, Satu Koskinen, Asta Laiho, Laura L. Elo, Mikael Knip, Riitta Lahesmaa, Alar Aints, Raivo Uibo
Summary: A study on the gene expression in B-cells from children developing CoD early in life revealed gene expression changes associated with CoD development, indicating the important role of B-cells in CoD development.
Article
Computer Science, Information Systems
Gunjan Chandra, Pekka Siirtola, Satu Tamminen, Mikael J. Knip, Riitta Veijola, Juha Roning
Summary: Clinical data analysis has the potential for breakthroughs, but the sensitive information in the data raises ethical concerns. Data anonymization is a common solution, but conventional methods have limitations. This study evaluates synthpop, an emerging tool for data anonymization, and finds that it successfully preserves utility and complexity, making it a valuable option for data sharing and protection.
Article
Microbiology
Joachim Johansen, Koji Atarashi, Yasumichi Arai, Nobuyoshi Hirose, Soren J. Sorensen, Tommi Vatanen, Mikael Knip, Kenya Honda, Ramnik J. Xavier, Simon Rasmussen, Damian R. Plichta
Summary: Distinct gut virome composition in centenarians, with higher diversity and lytic activity compared to younger and older individuals, suggests a potential role in healthy aging. Furthermore, phage-encoded sulfate metabolism genes in the centenarian gut microbiome may contribute to mucosal integrity and resistance to pathogens.
NATURE MICROBIOLOGY
(2023)
Editorial Material
Pediatrics
Mikael Knip
Article
Pediatrics
Kaija-Leena Kolho, Tapio Lahtiharju, Laura Merras-Salmio, Mikko P. Pakarinen, Mikael Knip
Summary: This study aims to investigate the characteristics of liver biochemistry in term infants. The study found that several analytes of liver biochemistry were higher than the currently used upper reference limits at 3 and 6 months of age in healthy term infants, and exclusively or partially breastfed infants showed higher values than formula-fed infants.
EUROPEAN JOURNAL OF PEDIATRICS
(2023)