Thursday, January 18, 2024, at 23h15 BE
HIBN, VFP, AAQC

2024 NEJMjw - Toothbrushing Lowers Risk for HAP (JAMA)
1. 𝙄𝙉𝘼𝘼𝙌𝘾 ᴮᴼ BRIEF: Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2023 (Dec 18), JAMAim, ? ➖ MA of 15 RCT ➕ 2800 ➕ ? ➖ PICO:
- P: adults in iMV
- I: toothbrushing and tongue brushing with chlorhexidine or plaque- removing toothpaste – 2-4x/day
- C: chlorhexidine swabbing – 2x/day
- pOC: HAP risks + MM
- sOC: dMV + ICU LOS
3. R ➩ ↓ HAP ➩ 33% ➕ ↓ MM 19% 🟰 ↓ dMV (1.2 d) ➕ ICU LOS (1.8d)4. Toothbrushing is not always easy for patients with endotracheal tubes, but these data support the effort to incorporate regular toothbrushing. (Paul Mueller)

2023 NEJMjw - Upper Gastrointestinal Bleeding in Patients Using Aspirin for Primary Prevention of Adverse CVE (Am J Med)
1. 𝙄𝙉𝘼𝘼𝙌𝘾 ᴮᴼ BRIEF: Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2023, AJM, USA ➖ retrospective OBS ➕ 535K (≥45yo) ➕ 2016-2020 ➖ PICO:
- P: middle-aged + older pxs
- I: ASA (primary prevention)
- C: ✖︎
- O: UGI bleeding INCIDENCE
3. R1 ➩ age- and sex-adjusted UGI bleeding ➩ INCIDENCE ↗️ 7.5 - 10 per 100k
4. R2 ➩ >75yo had ↗️ INCIDENCE + highest RATES
5. R3 ➩ all pxs ↗️ incidence
6. R4 ➩ UGI bleeding ≈ HIGHER ODDS OF: H admission + endoscopy + transfusioN.
7. The worrisome trend of increasing UGI bleeding should remind us to ask patients about aspirin use for primary prevention and to discourage its use for this purpose in people who are 60 or older. (Rahul Ganatra)
8. ASPREE + ARRIVE trials SHOWED THE SAME! (NEJM 2018, LANCET 2018)
atory settings and/or pharmacological or fluidic therapy were avoided either before or during TCD examination.

⏳ TIME MANAGEMENT

01:04:25
Round: 4  05:44:59 Q&A
Round: 3  16:11:29 Article 2
Round: 2  37:19:66 Article 1
Round: 1  05:09:75 Refresh

Thursday, January 22, 2024, at 23h15 BE
HIBN, AAQC

2023 LANCET - Seeing the humanity in health [ed]
1. Restricted numeric approach (statistics, averages and generalities) is different from the other elements concerning human beings and their realities
2. Human emotions should be very much considered to understand the faces behind health research (joy, fear, pain, hope, death)
3. Lancet shows images illustrating these scenes.
4. Nigeria example: universal health coverage + 2023 Wakley Prize Essay competition (lack of access to health care)
5. We are social animals, and personal stories help us to empathise. They make us care. (LANCET)
6. Care + compassion = COMMON THEMES. It gives some examples (kid helping in an Ebola centre, volunteers, etc.)
7. Our lives, and our health, are shaped by our physical, social, biological, and political surroundings. (LANCET)
8. Confronting challenges of humanity are the foundation of health research and advocacy.
9. Images + story telling are powerful.

2024 PSYADV - What Is the Impact of Shift Work on Sleep and Sleep Disorders Risk (FP)

DIS = disorders,

1. 𝙄𝙉𝘼𝘼𝙌𝘾 ᴮᴼ BS: Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2023, Front Psychiatry, BE ➖ survey ➕ >50k responded ➩ >37.5k included ➕ ? ➖ PICO:
* P: Belgian, >18yo (60% women)
* I: HSDQ applied
* C: none
* O: prevalence of sleep disorders + factors associated
3. R ➩ Day workers = > insomnia + movement DIS.
4. R ➩ Other shifts = insomnia + hypersomnia, and circadian rhythm sleep-wake disorders
5. R ➩ Other shifts 🆚 night shifts = ↗ prevalence (sleep DIS + sleep DIS comorbidity) (p<0.01) = ‘>1/2’ 1 sleep DIS = ‘>1/4’ ≥2 sleep DIS
6. 50% early morning, night or rotating shifts 🆚 25% regular day and evening shifts 🟰 SHORT SLEEP (p<0.01)7. Evening shifts 🆚 other shifts 🟰 LONG SLEEP (p<0.01)8. Male, elder, and lower education ≈ ↑ prevalence of SHORT sleep. 9. LIMITATIONS: selection bias ➕ use of HSDQ (polysomnography) ➕ small sample size10. Regular assessment of sleep quality and quantity and screening for disordered sleep in those working shifts might be crucial to timely treat sleep disorders

⏳ TIME MANAGEMENT:

46:40:80
Round: 6 00:17:08 Comments
Round: 5 26:53:06 Article 2
Round: 4 06:08:14 Wrap up
Round: 3 22:38:02 Article 1
Round: 2 02:56:78 Past JR
Round: 1 02:58:04 Articles decision

Thursday, January 25, 2024, at 18h25 BO
HIBN, JJFM, MACR, AAQC

2023 ICUmmp - A Very Old Patient in the ICU, Much More Than an Acute Organ Dysfunction (putowski)

CI = cognitive impairment, MOF = multi-organ failure
1. Ageing + shortage of ICU beds = demand of ICU supply of resources
2. Fluids are relevant in ELDERLY, DUE TO cardiovascular changes (aging changes described in Table 1)
3. Family support is PARAMOUNT considering ABCDEF bundle to avoid DELIRIUM
4. Cognitive impairment are 40% in elderly and it is difficult to measure in the ICU (though IQCODE exists)
5. Greater cognitive decline was associated with 30-day mortality.
6. ↑ sympatechic tone is RELEVANT in elderly patients
7. VIP-1, VIP-2 and the CLUSTER analysis ➩ one-point ↗ in SOFA = ↗ MM 30
8. ETHICA ➩ elderly were RELUCTANT to accept life-sustaining TTOs
9. mini–mental state examination (MMSE) = 30-point questionnaire = used to measure CI
10. Long-term OC after ICU are strongly determined by pre-ICU functional trajectories (Ferrante et al. 2015).
11. SOFA ➩ initially SEVERITY description ➩ later CORRELATED with MM ➩ described in VIP pxs with VIP-1, 2 and COVIP studies.
12. SOFA continues to be a globally utilised universal METHOD for MOF assessment.

2023 WHO - Preventing antimicrobial resistance together (vid)

AMR = antimicrobial resistance
1. AMR SHOULD be stopped
2. AMR ➩ contributes to ±5 million deaths / year
3. AMR = MISUSE and OVERUSE of ATBs
4. World AMR awareness week:
* Clean water
* Sanitation y hygiene
* INF prevention + control in health facilities
* VAX for children, elderly + at risk
* Responsible use of ATB
5. It is a shared responsibility
6. HIBN: avoid automedication
7. MACR: avoid contaminated water

⏳ TIME MANAGEMENT

46:40:80
Round: 4 05:18:29 Comments
Round: 3 57:41:22 Article 2
Round: 2 05:01:72 Article 1Round: 1 05:30:49 Last JR

Monday, January 29, 2024, at 18h29 BO
HIBN, JJFM, MACR, AAQC

2024 ICUmmp - Scientists discover first new ATB in six decades (Nature)

1. 𝙄𝙉𝘼𝘼𝙌𝘾 ᴮᴼ BS: Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2024, NATURE, USA ➖ trial (experimental) ➕ 39k compounds of ATB + 12M human body compounds ➕ ? ➖ PICO:
- P: mice and human MODELS
- I: ZOSURABALPIN for MRSA
- C: none
- O: effectivity to reduce the population
3. R ➩ it works, both in test tubes and mice (demonstrated exceptional efficacy)
4. Blocked the PROTECTIVE MEMBRANE by the transportation of a lipopolysaccharide, to the outer cell, ✖ ➩ formation of the protective membrane ➩ cell death.
5. 21 researchers involved
6. Long way to go before its potential use in hospitals
7. A. baumanni has zosurabalpin, in Phase I (Roche), effectively treats highly drug-resistant contemporary isolates of CRAB in vitro and in mouse models of infection. ➩ CRAB = Carbapenem-resistant acinetobacter baumanni

2023 SPARK - How to start an email_ The best and worst email greetings (R+R]

1. Types of greetings, which to use and NOT to use
2. In Spanish, it is mostly “good morning” (JMCM)
3. Context and recipient are important to set the email GREETING
4. TO USE
- Hi
- Hello
- Dear
- Dear TITLE
- Greetings
- Hi there
5. NOT TO USE
- Hey!
- Yo!
- HI nickname
- To whom it may concern
- Dear Sir/Madam
- Dear JOB TITLE
- Good morning/afternoon!
- Name
- Wrong or misspelled name
- No greeting

2024 LANCET - Virological characteristics of the SARS-CoV-2 JN.1 variant (kaku) [corr]

Mlm = multinomial logistic model
1. SARS-CoV-2 BA.2.86 main lineage BEFORE JN.1
2. BA.2.86 ➩ August 2023
3. ≠ Ω XBB lineages
4. JN.1 harbours Leu455Ser + 3 mutations in non-spike proteins
5. Leu455Phe ➩ contributes to increased transmissibility + immune escape
6. EFFECTIVE REPRODUCTIVE NUMBER of JN.1 was done ➩ FR, UK, ES ➩ 25 sequences of JN.1 with BAYESIAN mlm
7. JN.1 might become the dominant lineage worldwide ➩ INDEED in Nov 2023 OVERTOOK HK.3 in FR + ES
8. Infectivity = rodent sera infected OR immunized (w_BA.2.86) ➩ NT50 against JN.1 was similar to that against BA.2.86… ∑ Leu455Ser DOES NOT affect antigenicity of BA.2.86
9. In breakthrough INF sera w_XBB.1.5 + EG.5.1 against JN.1 was ↓ than HK.3 (NT50)
10. JN.1 shows robust resistance to monovalent XBB.1.5 vaccine COMPARED with BA.2.86
11. JN.1 is one of the most immune-evading variants to date

⏳ TIME MANAGEMENT:

01:23:09
Round: 7 05:33:04 Comments
Round: 6 17:32:89 JN.1. variant
Round: 5 28:34:06 SPARK, email greetings
Round: 4 19:39:50 ICU management and practice ART
Round: 3 05:50:59 Article decision
Round: 2 00:00:02 TEST
Round: 1 05:59:12 Refresh

Thursday, February 15, 2024, at 18h55 BO
HIBN, AAQC

2023 HEALIO - A look at intermittent fasting. A potential strategy for pxs w_ obesity (Obesity Pillars)

1. Common IF pattens:

    - Daily time-restricted feeding
   - OMAD plan
   - 5:2 plan
2. Calorie restriction is relevant to support IF
3. Hypoglycemia should be prevented and anticipated
4. Protein intake is PARAMOUNT to apply safely IF
5. Benefits: improvement in hunger, weight loss, insulins sensitivity, blood pressure, inflammation, stress response and sleep. Reduces decision fatigue. Useful for shift workers ➩ organize meals in a consisten time frame.
6. Risks: hypoglycemia, SARCOPENIA, dehydration. AVOID PROLONGED fasting ➩ ♥️ dysrhythmias, gout, kidney stones and postural hypotension.
7. Application should be:   * First, ensure protein intake   * Second, ↓ carbohydrate intake   * Apply IF8. Always combine nutrition (food QUALITY MATTERS) and lifestyle changes.9. The 2022 Clinical Practice Statement on Nutrition and Physical Activity by the Obesity Medicine Association would be a valuable starting place for a clinician interested in using IF with pxs.

Thursday, February 01, 2024, at 18h29 BO
HIBN, AAQC

2023 NEJM - Notable Articles of 2023 (NEJMed)

1. ARTs of interest:
   - Semaglutide in pxs with HFpEF & 🐷
   - Dupilumab for COPD with T2 Inflammation indicated by Eosinophil Counts
   - RCT of Early Detection and TTO of Postpartum Hemorrhage
   - Ketamine versus ECT for Nonpsychotic Treatment-Resistant Major Depression
   - Inaxaplin for Proteinuric Kidney Disease in Persons with Two APOL1 Variants
   - Treatment Strategy for Rifampin-Susceptible Tuberculosis
   - Phase 2 Trial of Baxdrostat for TTO-Resistant HTA
   - Empagliflozin in pxs with CKD
   - Lecanemab in Early Alzheimer’s Disease

2. LECANEMAN in ALZHEIMER
CDR-SB = Clinical Dementia Rating–Sum of Boxes, eALZ = early Alzheimer disease, AdvEve = adverse events
   - 2023, NEJM, USA ➖ ph3_mc_db_RCT ➕ >1500pxs ➕ 18m ➖ PICO:
      * P = early alzheimer pxs (50-90yo) w_amyloid on PET OR CSF testing      * I = lecanemab      * C = placebo      * pOC = change of CDR-SB at 18m      * R = less reduction in I   - CDR-SB = ranges from 0 to 18 ➩ ↗️ scores = GREATER impairment   - Similar AdvEve in I + C ➩ most common = infusion-related reactions ➕ amyloid-related imaging abnormalities w_ edema OR effusions.   - Lecanemab ↓ markers of amyloid in eALZ ➕ mod_less ↘️ on measures of cognition and function at 18m BUT was ≈ adverse events.   - Study done during C19

Thursday, March 28, 2024, at 18h29 BO
HIBN, AAQC

2024 NPR - In a pandemic milestone, the NIH ends guidance on C19 TTO (Huang) [r].pdf

PhysiIDSA = Infectious Diseases Society of America, ACP = American Colleague of Physicians, GL = guidelines, Ab = antibodies
L=s, Ab = antibodies
1. NIH will end their guidance. They recommend to follow IDSA ➕ ACP
2. Last 4 years NIH has been publishing dynamically the GL
3. 50M times were visited the NIH GL
4. TURNING POINTS:
     - 2020 ➩ first pandemic wave ➩ 1st GL “we don’t know what does and does not work”
     - June 2020 ➩ steroids (dexamehasone)… combination with antivirals
     - A year into the pandemic ➩ lab-made Ab ➩ monoclonal Abs gave so much insight into the virus itself ➩ challenged by fast-changing spike protein (new strains)
     - End 2021 ➩ Molnupiravir (Merck) ➕ Paxlovid (Pfizer) ➩ warts (limitations) = molnupiravir (not SO effective), paxlovid (many interactions e.g. statins)
5. Underuse ➩ Molnupiravir 2% and Paxlovid 15% ➩ of C19 eligible pxs
6. Study 2021-2022 ➩ federal government bought PAXLOVID + MOLNUPIRAVIR ➩ then transitioned to commercial market ∑ less use. (Have likely further declined since late 2023)7. Patient dilema ➩ pxs do not BELIEVE they will get very sick (risk factors assessment)8. We need PATIENT EDUCATION ➩ how the drug work ➕ when they’re most effective 9. What’s in the pipeline? Japanese company - SHIONOGI ➩ pill course for ACUTE ➕ long COVID

Monday, April 18, 2024, at 17h15 BE 
HIBN, AAQC

2024 MB - Trials of Psychological Interventions for Depression Rarely Assess Suicidality (LANCETpsy).pdf
1. Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2024, Lancet PSY, NL ➖ srMA of RCTs➕468 RCT ➕ up to May 2023 ➖ PICO:
     * P: RCTs
     * I: studies w_depression + suicidality + psy_INT
     * C: NA
     * O: psy_INT were efficient?
3. R:
   - aggregated analysis = suicidal ideation (SI) ➕ suicide risk (SR) = ↘️ immediately af_I (SMD,    -0.31; 95% CI, -0.60 to -0.03; I2, 68%).
   - I ↘️ SI + SR af_I 🆚 active control groups (SMD, -0.34; 95% CI, -0.66 to -0.02; I2, 72%) 🆚 tto as usual as the control (SMD, -0.40; 95% CI, -0.74 to -0.05; I2, 71%)
   - SI alone ➩ the same ↘️ (SMD, -0.36; 95% CI, -0.70 to -0.02; I2, 70%)
   - At f-up ➩ no effects (SMD, -0.49; 95% CI, -1.31 to 0.32; I2, 79%)
4. No significant effect in studies wo_
1. Direct intervention,
2. Single-item measures,
3. Continuous outcomes.
5. Assessment of suicide was rarely reported in trials of psychological interventions targeting depression. (authors)
6. ‘Monitoring and reporting of suicidal thoughts’ need ↗️ - even ‘risk of suicide’ is excluded.”
7. Limitations:
   - Small sample
   - Single-item suicide OC
   - Combination of ≠ suicide OC

2024 PSYADV - ADHD and Loneliness Affects Mental Health in Young People (JAD).pdf
1. 2024, J. Atten. Disord, ? ➖ srMA (cross-sectional or longitudinal quantitative) ➕ 20 studies (1253 pxs) ➕ ? ➖ PICO:
    - P: young adults (10-24 years) w_ADHD
    - I: loneliness w_ADHD
    - C: wo_ADHD
    - O: prevalence of loneliness + effect of loneliness on mental health
2. Methods:
    - 1 measure of LONE
    - ADHD was verified via Diagnostic and Statistical Manual of Mental Disorders (DSM) diagnoses and ICD
3. Results ➩
    - 9/17 ↑ levels of LONE
    - 4/17 no diff
    - 3 wo_conduct significance test
    - 1 ↓ levels of LONE
4. Meta ➩
    - P = ↑ levels of LONE (Hedges g, 0.41; 95% CI, 0.25, 0.58; P <.001)
    - No sig influence of SIZE despite concerns of POTENTIAL PUBLICATION BIAS (tested by moderator analysis)
5. sr ➩
    - LONE ≈ mental health difficulties in P
    - ⊕ LONE ≈ externalizing behavior, internalizing behaviors, depression, anxiety, and addiction
    - ADHD is a sig PREDICTOR of major depressive disorder onset (confounder were controlled)
6. This review highlights that loneliness may be an important problem in ADHD and clinicians should be aware of and assess the potential for elevated loneliness in this population.
7. Limitations ➩
    - Small n of studies
    - Narrow focus on LONE
    - Dichotomous categorization of ADHD (less severe excluded?)

Monday, April 15, 2024, at 17h15 BE 
HIBN, AAQC

Is OCD Linked to a Higher Risk for Death From Natural and Unnatural Causes
1. Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2024, BMJ, SE ➖ matched coacher + sibling cohort ➕ 15M (selected 61k + 34k ) ➕ Jan 1973 to Dec 2020 (48y) ➖ PICO:
     * P: >6y SE people
     * I: OCD dx (affected)
     * C: unaffected overall and siblings
     * pOC: all-cause + cause specific
3. R ➩ adjusted stratified cox proportional:
     * All cause = HR, 1.82
     * Natural causes = HR, 1.3
     * Unnatural causes = HR, 3.3
4. Diagnosed pxs, more likely:
     * SE natives
     * Less educated
     * Single
     * Low income
     * Psychiatric disorders
5. Risk factor for specific causes:
     * endocrine, nutritional, and metabolic diseases;
     * mental and behavioral disorders (dementia, vascular dementia, and mental and behavioral disorders alcohol related)
     * nervous, circulatory, respiratory, digestive, and genitourinary systems
6. Less risk in pxs NEOPLASMS
7. Sibling cohort ➩ 8y ➕ more likely women + same as point 4 ➕ MM w_OCD 4.7 + MM wo_OCD 2.7 ➕ adjusted risks acMM 1.8, natural 1.5, unnatural 3.1.
8. Limitations ➩ selection bias due to:
     * data source constraints
     * underpowered estimates for specOC
     * mediator effects of covariates
     * lack of info on lifestyle factors * uncertainty ‘generalizability to diverse populations’ * uncertainty ‘HC settings outside of Sweden’

Trials of Psychological Interventions for Depression Rarely Assess Suicidality
1. Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2024, Lancet PSY, NL ➖ srMA of RCTs➕468 RCT ➕ up to May 2023 ➖ PICO:
* P: RCTs
* I: studies w_depression + suicidality + psy_INT
* C: NA
* O: psy_INT were efficient?
… to continue…

Thursday, April 25, 2024, at 17h15 BE 
HIBN, AAQC

   - GPT-4 ↗️ than physicians (PSY) ➕ similarly to physician (SX & IM)
   - GPT-4 ↘️ in PED & OB/GYN ➕ ↗️ than physicians
   - GPT-3.5 did not pass the examination in any discipline ➕ ↘️ to physicians (5 disciplines)
   - GPT-4 passed 4 of 5 (score > 65%)
6. Strengths:
   - Large cohort allowing direct comparison ↔ GPT models
   - Open sharing of the examinations
   - ⊕ builds confidence in the technology
   - ⊕ too early to integrate in clinical practice
   - ⊕ good for medical education, simulations, personal assessment, and feedback
   - 🔍 combination of physician + AT = ↗️ results 🆚 2 physicians 🆚 AI alone ➩ synergy matters
7. Limitations:
   - Image-based question exclusion
   - Linguistic ⌄ cultural biases due to translation
8. ∑ ➩ This juncture represents an opportunity to reshape physician training and capabilities in tandem with the advancements in AI.

⏳ TIME MANAGEMENT

01:32:45

Round: 3 00:18:09 Brief
Round: 2 01:29:52 continue ART
Round: 1 02:35:32 Last JR

2024 NEJMai - GPT versus Resident Physicians, A Benchmark Based on Official Board Scores (katz) [R].pdf
1. Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2024, NEJMai, IS ➖ retrospective cohort ➕ 849 physicians ➕ 2022 ➖ PICO:
   - P: medical specialists
   - I: GPT-3.5 & GPT-4
   - C: board residency examinations
   - O: PERFORMANCE of physician, GPT-4, GPT-3.5
3. Importance:
   - How LLM is relevant for medical education and clinical practice?
   - How different version of LLM perform?
   - 🔍 LLM performance with questions from simulated models (MedQA, MedMCQA, and MultiMedQA)
   - 🔍 LLM evaluated in real examination settings
4. Methods:
   - IN. Number of physicians ➕ scores
   - EX. Questions w_images
   - Translated from Hebrew to English
   - Run by standard Web application (OpenAI)
   - Minimize biases = ChatGPT refreshed between questions.
   - Two columns (excel) question and solution (4 mulitiple options)5. Results: IM = internal medicine, SX = surgery, OB/GYN = obstetrics and gynecology, PSY = psychiatry, PED = pediatrics

Monday, April 22, 2024, at 17h15 BE 
HIBN, AAQC

⏳ TIME MANAGEMENT

49:01:36

Round: 4 17:09:46 ART 2 selection

Round: 3 20:29:65 ART 1

Round: 2 09:04:00 ART selection + ART 1

Round: 1 02:18:24 Past JR

2024 NEJMai - GPT versus Resident Physicians, A Benchmark Based on Official Board Scores (katz) [R].pdf
1. Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2024, NEJMai, IS ➖ cohort ➕ 849pxs ➕ 2022 ➖ PICO:
    - P: medical specialists
    - I: GPT-3.5 & GPT-4    - C: board residency examinations    - O: PERFORMANCE3. To continue…

2024 AIH - 10 things you may have suspected about AI but didn’t know for sure till now (pearson).pdf
0. Ten points by analysts of Stanford University
1. The public is pessimistic about AI’s economic impact.
    - Survey ➩ 37% AI will ↗️ job
    - 32% will ↗️ job market
    - 3 of 10 ONLY think it is positive for the economy (HIBN)
2. Demographic differences exist regarding AI optimism.
    - Gen Z more OPTIMISTS than baby boomers
    - Higher incomes + education levels = ↗️ optimists (entertainment, health, economy)
3. ChatGPT is widely known and widely used.
    - Univ. Toronto ➩ 63% aware ➕ 50% of them USE it
4. AI helps medicine take significant strides forward.
    - 2023 SEVERAL medial systems were launched
       * EVEscape = predicts pandemics
       * AlphaMissence = mutation classification
5. Highly knowledgeable medical AI has arrived.
    - ↗️ MedQA benchmark = to assess AIs clinical knowledge (90% accuracy)
    - Since 2019 was 3X
6. The FDA is approving more and more AI-equipped medical devices.
    - 139 devices
    - Since 2012 = 45x more
7. Robust and standardized evaluations for GenAI responsibility are seriously lacking.
    - Lack of standardization in REPORTING
    - Complicates risks & limitations comparisons
8. Researchers have discovered more complex vulnerabilities in large language models.
    - Security weaknesses (adversarial prompts)
    - Less ovbious strategies ➩ repeat words infinitely
9. The number of AI incidents continues to rise.
    - 123 incidentes in 2023
    - 20x ↗️ since 2013
    - Sexually explicit deepfakes
10. ChatGPT is politically biased.
    - Towards democrats in the US
    - Towards Labour Party in UK

Monday, May 6, 2024 at 17h45 BEHIBN, AAQC

2024 NN - Thought Provoking Work May Reduce Later Life Cognitive Decline (AAN - NEUROLOGY).pdf

Glossary: MCI = mild cognitive impairment, ↓r = may lower risk, RTI = routine task intensity

1. Complex, thought-provoking work ↓r MCI
2. Y, J, C ➖ T ➕ N ➕ t ➖ PICO: ➩ 𝙄𝙉𝘼𝘼𝙌𝘾 ᴮᴼ
3. 2024, NEUROLOGY, Norway (Oslo U.) ➖ cohort (30a - 65a) ➕ 7k ➕ 2a (2017-19) ➖ PICO:
   - P: people working
   - I:  registry-based trajectories (305 occupations) + dementia diagnosis
   - C: NA
   - O: correlation
4. Methods:
   - Nature of tasks = routine manual, routine cognitive, non-routine analytical, and non-routine interpersonal
   - Non-routine analytical tasks = activities that involve analyzing information, engaging in creative thinking and interpreting information for others.
   - Non-routine interpersonal tasks = establishing and maintaining personal relationships, motivating others and coaching.
   - Non-routine cognitive = public relations and computer programing.
5. Additional data. Participants were divided into 4 groups based on the degree of cognitive stimulation:
   - Low RTI: Primary education teaching associate professionals (44.3%), secondary education teaching professionals (9.9%), other public service administrative professionals (5.6%), crop and animal producers (5.6%), and college, university, and higher education teaching professionals (4.2%).
   - Intermediate-low RTI: Child care workers (11.1%), Nurses (10.7%), other public service administrative associate professionals (5.3%), bank associate professionals (4.4%), and other public service administrative professionals (4.4%).
   - Intermediate-high RTI: Shop salespersons and other salespersons (retail) (17.6%), nursing assistants and care assistants (12.9%), personal care and related workers not elsewhere classified (7.3%), secretaries (5.5%), and caretakers (3.8%).
   - High RTI: Helpers and cleaners in offices and other establishments (9.7%), mail carriers and sorting clerks (7.0%), road workers and construction workers (5.1%), paper-pulp and papermaking plant operators (3.8%), and heavy truck and lorry drivers (3.5%).
6. Results:
   - Adjustments (demographic & lifestyle) STILL showed a 66% ↑r MCI in ‘least cognitively demanding jobs’.
   - Participants in the highest cognitive demand jobs had ↓ MCI (27%) 🆚 lowest demand jobs (42%).
   - The most common job (highest cognitive demands) = teaching.    - The most common jobs (lowest cognitive demands) = mail carriers and custodians.7. HUNT4 70+ study8. CAUTION ➩ It only shows an association.

⏳ TIME MANAGEMENT

37:00:25

Round: 2 34:11:60 ART 1
Round: 1 02:48:65 Last JR

Thursday, May 2, 2024, at 17h15 BE 
HIBN, RYCC, AAQC

2024 ICM - Noninvasive neuromonitoring in acute brain injured patients (brasil) [ED] 

NIM = noninvasive monitoring, BUS = brain ultrasound, TCCD = transcranial color-duplex sonography, CBF = cerebral blood flow, ƒ-up = follow-up, PI = pulsatility index, QEEG = quantitative EEG

1. See past notes…
2. Electrical activity
   - EEG ➩ for comatose
   - Detects seizures + treatment starting decision (and escalating)
   - Structural brain damage + 🧠 dysfunction ≈ seizures
   - QEEG: digital signals analyzed w_mathematical algorithms (frequency bands)
   - Non-convulsive status detection
   - Also detects ischemia, bleeding, hydrocephalus, brain swelling or herniation and to assess sedation depth
   - Malignant patterns ➩ poor OC
3. Function
   - PUPILLOMETER ➩ delivers 3 sec flash of light + optimizes accuracy
   - Objective, quantitative, and reliably repeatable assessment of the brainstem function
   - Evaluates central reflex pathways
   - NPi good prognostic tool ➩ post CA + NCC pxs ➩ ≈ ↑ ICP when abnormal (0-3)
4. Mechanical properties
   - Volume matters (intracranial compliance - intracerebral compliance) to CEREBRAL blood perfusion
   - B4C ➩ nanometric resolution of pulsatile elastic movement within heartbeats. Shows ICP surrogate waveforms in real time.
   - P2/P1 ratio and time-to-peak ➩ assess ICC deterioration
   - It has potential utilization in clinics IF ICC is compromised.
   - NIM synergism = BUS + B4C = diagnostic power enhancement
   - Optic nerve sheath diameter ➩ BUS ➩ if CSF pressure ↑, then >5.8mm ≈ ↑ ICP
5. Oxygenation
   - NIRS ➩ rsO2 = Hb O2 saturation
   - Two sensors over the frontal lobes
   - Balance: DO2 ↔ VO2
   - rSO2 < 50% ⌄ ↓ >10-20% from basal   - Technical issues ➩ contamination by extraCRANIAL signals

6. Take-home message   

   - ADVANTAGES: safe, low cost, bedside, repeatable   

   - LIMITATIONS: accuracy, need of training.

   - Cannot SUBSTITUTE invasive methods.

2024 FRONTIERS - Cake and cookies may increase Alzheimer’s risk (FRONTIER nutrition).pdf

1. UPF = ultraprocessed food
2. Linked with dementia
3. UPF ➩ frozen meals, soft drinks, cake mixes, and fast food – added sugars or fats ➕ artificial colors and flavors.4. Frontiers in Nutrition, BR ➩ UPF ≈ Alzheimer ➩ most, but one with DEMENTIA5. Caution to interpret

2024 NEJMjw - How Does Lifestyle Affect Cognition and Brain Pathology (JAMA)

RMAP = Rush Memory and Aging Project

1. Y, J, C ➖ T ➕ N ➕ t ➖ PICO:
2. 2024, JAMA NEUROLOGY, ? ➖ clinicoPATHOL cohort ➕ 586 ➕ ? ➖ PICO:
   - P: postmortem brain pathology RMAP
   - I: lifestyle factors
   - C: NA
   - O: global cognitive score
3. Healthy lifestyle score = self-reported nonsmoking status + physical activity of ≥150 minutes per week + light-to-moderate alcohol consumption + adherence to the Mediterranean-DASH y (MIND diet), cognitive activity.
4. Score = 0-5 = higher WAS healthier lifestyle5. Results: - Lifestyle ➩ higher scores were healthier - Dementia-related brain pathologies = ß-amiloid load, phosphorylated tau tangles… and other6. Patients were studied close to death, 1y between cognitive test and death7. Better global cognitive performance = ↑ lifestyle scores = ↓ ß-amyloid load (12% of ‘lifestyle - cognitive performance’ = ß-amyloid load)

⏳ TIME MANAGEMENT

56:40:95

Round: 4 21:20:74 ART 3
Round: 3 12:49:19 ART 2

Round: 2 19:27:72 ART 1

Round: 1 03:03:30 past JR