Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing.
- MeSH
- Big Data * MeSH
- Glioblastoma * MeSH
- Humans MeSH
- Information Dissemination MeSH
- Machine Learning MeSH
- Rare Diseases MeSH
- Check Tag
- Humans MeSH
- Publication type
- Journal Article MeSH
- Research Support, N.I.H., Extramural MeSH
We investigated and evaluated post-irradiation survival in cyclooxygenase-2-deficient (COX-2 KO) mice. Thirty-day survival following exposure of COX-2 KO mice to a lethal dose of 8.5 Gy of gamma-rays was observed to be statistically significantly lower in both males and females, as well as when the sexes were merged, in comparisons with their wild-type counterparts. These findings were related to the previous observations concerning the detrimental influence of the COX-2 genetic disruption on hematopoiesis in sublethally irradiated mice. Deteriorated post-irradiation survival of COX-2 KO mice confirmed the previously anticipated conclusion regarding negative influence of the antiinflammatory action of COX-2 deficiency under the conditions of exposure of the animals to ionizing radiation.
- MeSH
- Cyclooxygenase 2 deficiency radiation effects MeSH
- Survival Rate trends MeSH
- Mice, Knockout MeSH
- Mice MeSH
- Gamma Rays adverse effects MeSH
- Animals MeSH
- Check Tag
- Male MeSH
- Mice MeSH
- Female MeSH
- Animals MeSH
- Publication type
- Journal Article MeSH
We investigated hematopoiesis in untreated and ionizing radiation-exposed cyclooxygenase-2-deficient (COX-2 KO) mice. We performed a complex hematological analysis of 16 parameters in untreated COX-2 KO male mice or COX-2 KO male mice irradiated with the dose of 4 Gy of gamma-rays and their wildtype littermates. At baseline, hematopoiesis was increased in COX-2-deficient mice, but attenuated by irradation in COX-2-deficient mice compared to wildtype. To conclude, the anti-inflammatory action of the COX-2 genetic disruption plays a positive role in hematopoiesis under basal conditions but is detrimental following radiation exposure.
- MeSH
- Cyclooxygenase 2 deficiency radiation effects MeSH
- Hematopoiesis physiology radiation effects MeSH
- Radiation, Ionizing * MeSH
- Mice, 129 Strain MeSH
- Mice, Knockout MeSH
- Mice MeSH
- Animals MeSH
- Check Tag
- Male MeSH
- Mice MeSH
- Animals MeSH
- Publication type
- Journal Article MeSH
... Filipovič GONČARENKO 38 -- Boris Alexandrovič GREKOV 39 -- Jekaterina GRIVNOVÁ 40 -- Sergej Sergejevič GRUZDEV ...
1. vydanie 75 stran : portréty, faksimile ; 24 cm
- MeSH
- History, 20th Century MeSH
- History of Medicine MeSH
- Emigrants and Immigrants MeSH
- Physicians MeSH
- Check Tag
- History, 20th Century MeSH
- Publication type
- Bibliography MeSH
- Biography MeSH
- Geographicals
- Czechoslovakia MeSH
- Conspectus
- Patologie. Klinická medicína
- Biografie
- NML Fields
- dějiny lékařství
- humanitní vědy a umění
- NML Publication type
- studie
