An increasing volume of prostate biopsies and a worldwide shortage of urological pathologists puts a strain on pathology departments. Additionally, the high intra-observer and inter-observer ...variability in grading can result in overtreatment and undertreatment of prostate cancer. To alleviate these problems, we aimed to develop an artificial intelligence (AI) system with clinically acceptable accuracy for prostate cancer detection, localisation, and Gleason grading.
We digitised 6682 slides from needle core biopsies from 976 randomly selected participants aged 50–69 in the Swedish prospective and population-based STHLM3 diagnostic study done between May 28, 2012, and Dec 30, 2014 (ISRCTN84445406), and another 271 from 93 men from outside the study. The resulting images were used to train deep neural networks for assessment of prostate biopsies. The networks were evaluated by predicting the presence, extent, and Gleason grade of malignant tissue for an independent test dataset comprising 1631 biopsies from 246 men from STHLM3 and an external validation dataset of 330 biopsies from 73 men. We also evaluated grading performance on 87 biopsies individually graded by 23 experienced urological pathologists from the International Society of Urological Pathology. We assessed discriminatory performance by receiver operating characteristics and tumour extent predictions by correlating predicted cancer length against measurements by the reporting pathologist. We quantified the concordance between grades assigned by the AI system and the expert urological pathologists using Cohen's kappa.
The AI achieved an area under the receiver operating characteristics curve of 0·997 (95% CI 0·994–0·999) for distinguishing between benign (n=910) and malignant (n=721) biopsy cores on the independent test dataset and 0·986 (0·972–0·996) on the external validation dataset (benign n=108, malignant n=222). The correlation between cancer length predicted by the AI and assigned by the reporting pathologist was 0·96 (95% CI 0·95–0·97) for the independent test dataset and 0·87 (0·84–0·90) for the external validation dataset. For assigning Gleason grades, the AI achieved a mean pairwise kappa of 0·62, which was within the range of the corresponding values for the expert pathologists (0·60–0·73).
An AI system can be trained to detect and grade cancer in prostate needle biopsy samples at a ranking comparable to that of international experts in prostate pathology. Clinical application could reduce pathology workload by reducing the assessment of benign biopsies and by automating the task of measuring cancer length in positive biopsy cores. An AI system with expert-level grading performance might contribute a second opinion, aid in standardising grading, and provide pathology expertise in parts of the world where it does not exist.
Swedish Research Council, Swedish Cancer Society, Swedish eScience Research Center, EIT Health.
Purpose
To evaluate clinical variables, including magnetic resonance imaging (MRI) predictive of adverse pathology (AP) at radical prostatectomy (RP) in men initially enrolled in active surveillance ...(AS).
Methods
A population-based cohort study of men diagnosed with low-risk prostate cancer (PCa), in Stockholm County, Sweden, during 2008–2017 enrolled in AS their intended primary treatment followed by RP. AP was defined as ISUP grade group ≥ 3 and/or pT-stage ≥ T3. Association between clinical variables at diagnosis and time to AP was evaluated using Cox regression and multivariate logistic regression to evaluate the association between AP and clinical variables at last biopsy before RP.
Results
In a cohort of 6021 patients with low-risk PCa, 3116 were selected for AS and 216 underwent RP. Follow-up was 10 years, with a median time on AS of 23 months. 37.7% of patients had AP at RP. Clinical T-stage Hazard ratio (HR): 1.81, 95% confidence interval (CI) 1.04–3.18 and PSA (HR: 1.31, 95% CI 1.17–1.46) at diagnosis and age Odds Ratio (OR): 1.09, 95% CI 1.02–1.18), PSA (OR: 1.22, 95% CI 1.07–1.41), and PI-RADS (OR 1.66, 95% CI 1.11–2.55) at last re-biopsy were significantly associated with AP.
Conclusion
PI-RADS score is significantly associated with AP at RP and support current guidelines recommending MRI before enrollment in AS. Furthermore, age, cT-stage, and PSA are significantly associated with AP.
How people assess their social environments plays a central role in how they evaluate their life circumstances. Using a large probabilistic national sample, we investigated how accurately people ...estimate characteristics of the general population. For most characteristics, people seemed to underestimate the quality of others' lives and showed apparent self-enhancement, but for some characteristics, they seemed to overestimate the quality of others' lives and showed apparent self-depreciation. In addition, people who were worse off appeared to enhance their social position more than those who were better off. We demonstrated that these effects can be explained by a simple social-sampling model. According to the model, people infer how others are doing by sampling from their own immediate social environments. Interplay of these sampling processes and the specific structure of social environments leads to the apparent biases. The model predicts the empirical results better than alternative accounts and highlights the importance of considering environmental structure when studying human cognition.
Three billion people in low- and middle-income countries are exposed to household air pollution as they use biomass fuel for cooking. We investigated the associations between solid fuel use and ...nasopharyngeal (NP) inflammation, as well as the associations between high pneumococcal density and NP inflammation, in mothers and children in rural and urban Ethiopia.
Sixty pairs of mothers (median age, 30 years; range, 19-45 years) with a child (median age, 9 months; range, 1-24 months) were included from rural Butajira (n = 30) and urban Addis Ababa (n = 30) in Ethiopia. The cohort was randomly selected from a previous study of 545 mother/child pairs included 2016. Questionnaire-based data were collected which included fuel type used (solid: wood, charcoal, dung or crop waste; cleaner: electricity, liquefied petroleum gas). Nasopharyngeal (NP) samples were collected from all mothers and children and analyzed for the levels of 18 cytokines using a Luminex immunoassay. Pneumococcal DNA densities were measured by a real-time multiplex PCR and a high pneumococcal density was defined as a cyclic threshold (Ct) value ≤ 30.
Mothers from rural areas had higher median CXCL8 levels in NP secretions than those from urban areas (8000 versus 1900 pg/mL; p < 0.01), while rural children had slightly higher IL-10 levels than those from the urban area (26 vs 13 pg/mL; p = 0.04). No associations between fuel type and cytokine levels were found. However, a high pneumococcal density was associated with higher levels of cytokines in both mothers (CCL4, CXCL8, IL-1β, IL-6 and VEGF-A) and children (CCL4, CXCL8, IL-1β, IL-6 and IL-18).
No significant associations were found between solid fuel use and NP inflammation in Ethiopian mothers and children, but the inflammatory activity was higher in individuals living in the rural compared to the urban area. In addition, high cytokine levels were associated with high pneumococcal density in both mothers and children, indicating a significant impact of NP pathogens on inflammatory mediator levels in upper airways.
Social categorizations divide people into "us" and "them", often along continuous attributes such as political ideology or skin color. This division results in both positive consequences, such as a ...sense of community, and negative ones, such as group conflict. Further, individuals in the middle of the spectrum can fall through the cracks of this categorization process and are seen as out-group by individuals on either side of the spectrum, becoming inbetweeners. Here, we propose a quantitative, dynamical-system model that studies the joint influence of cognitive and social processes. We model where two social groups draw the boundaries between "us" and 'them" on a continuous attribute. Our model predicts that both groups tend to draw a more restrictive boundary than the middle of the spectrum. As a result, each group sees the individuals in the middle of the attribute space as an out-group. We test this prediction using U.S. political survey data on how political independents are perceived by registered party members as well as existing experiments on the perception of racially ambiguous faces, and find support.
Unreliable predictions can occur when an artificial intelligence (AI) system is presented with data it has not been exposed to during training. We demonstrate the use of conformal prediction to ...detect unreliable predictions, using histopathological diagnosis and grading of prostate biopsies as example. We digitized 7788 prostate biopsies from 1192 men in the STHLM3 diagnostic study, used for training, and 3059 biopsies from 676 men used for testing. With conformal prediction, 1 in 794 (0.1%) predictions is incorrect for cancer diagnosis (compared to 14 errors 2% without conformal prediction) while 175 (22%) of the predictions are flagged as unreliable when the AI-system is presented with new data from the same lab and scanner that it was trained on. Conformal prediction could with small samples (N = 49 for external scanner, N = 10 for external lab and scanner, and N = 12 for external lab, scanner and pathology assessment) detect systematic differences in external data leading to worse predictive performance. The AI-system with conformal prediction commits 3 (2%) errors for cancer detection in cases of atypical prostate tissue compared to 44 (25%) without conformal prediction, while the system flags 143 (80%) unreliable predictions. We conclude that conformal prediction can increase patient safety of AI-systems.
Carbon footprints-the greenhouse gas (GHG) emissions associated with consumer food choices-substantially contribute to climate change. Life cycle analyses from climate and environmental sciences have ...identified effective rules for reducing these food-related GHG emissions, including eating seasonal produce and replacing dairy and red meat with plant-based products. In a national UK survey, we studied how many and which rules our participants generated for reducing GHG emissions of produce, dairy, and protein-rich products. We also asked participants to estimate GHG emission reductions associated with pre-selected rules, expressed in either grams or percentages. We found that participants generated few and relatively less effective rules, including ambiguous ones like 'Buy local'. Furthermore, participants' numerical estimates of pre-selected rules were less accurate when they assessed GHG emission reductions in grams rather than in percentages. Findings suggest a need for communicating fewer rules in percentages, for informing consumers about reducing food-related GHG emissions.
Mapping the Structure of Semantic Memory Morais, Ana Sofia; Olsson, Henrik; Schooler, Lael J.
Cognitive science,
January/February 2013, Letnik:
37, Številka:
1
Journal Article
Recenzirano
Odprti dostop
Aggregating snippets from the semantic memories of many individuals may not yield a good map of an individual’s semantic memory. The authors analyze the structure of semantic networks that they ...sampled from individuals through a new snowball sampling paradigm during approximately 6 weeks of 1‐hr daily sessions. The semantic networks of individuals have a small‐world structure with short distances between words and high clustering. The distribution of links follows a power law truncated by an exponential cutoff, meaning that most words are poorly connected and a minority of words has a high, although bounded, number of connections. Existing aggregate networks mirror the individual link distributions, and so they are not scale‐free, as has been previously assumed; still, there are properties of individual structure that the aggregate networks do not reflect. A simulation of the new sampling process suggests that it can uncover the true structure of an individual’s semantic memory.
To investigate associations between depression, anxiety, and antidepressants before in vitro fertilization (IVF) and IVF cycle outcomes, including pregnancy, live birth, and miscarriage.
Nationwide ...register-based cohort study.
Not applicable.
Nulliparous women undergoing their first IVF cycle recorded in the Swedish Quality Register of Assisted Reproduction, January 2007 to December 2012 (n = 23,557).
Not applicable.
Associations between diagnoses of depression/anxiety, antidepressants, and IVF cycle outcome evaluated using logistic regression to produce adjusted odds ratios (AOR) and 95% confidence intervals (CI).
In total, 4.4% of women had been diagnosed with depression/anxiety and/or dispensed antidepressants before their IVF first cycle. The odds for pregnancy and live birth were decreased (n = 1,044; AOR = 0.86; 95% CI, 0.75-0.98; and AOR = 0.83; 95% CI, 0.72-0.96, respectively). For women with a prescription for a selective serotonin reuptake inhibitor (SSRI) only (n = 829), no statistically significant associations were found. Women with non-SSRI antidepressants (n = 52) were at reduced odds of pregnancy (AOR = 0.41; 95% CI, 0.21-0.80) and live birth (AOR = 0.27; 95% CI, 0.11-0.68). Women with a depression/anxiety diagnosis with no antidepressant (n = 164) also had reduced odds of pregnancy (AOR = 0.58; 95% CI, 0.41-0.82) and live birth (AOR = 0.60; 95% CI, 0.41-0.89). Among the women who became pregnant (39.7%), there were no statistically significant associations between exposure and miscarriage except for the women taking non-SSRI antidepressants (AOR = 3.56; 95% CI, 1.06-11.9).
A diagnosis of depression/anxiety and/or treatment with antidepressants before IVF was associated with slightly reduced odds of pregnancy and live birth. Women with the presence of depression/anxiety without antidepressants had a more pronounced reduction in odds, implying that the underlying disorder is important for the observed association.