แหล่งข้อมูลสุขภาพและวิทยาศาสตร์เบื้องหลัง Wellness Project

การประมาณแคลอรี & สารอาหารหลัก (บันทึกอาหาร)

When you log food (by typing a description, chatting with an AI coach, scanning a photo, or saving a recipe), the app's AI model estimates calories, protein, carbohydrates, fat, alcohol, and micronutrients for the foods and portions described. These are model estimates, not measurements, always marked as such, and you can edit every value before or after saving.

• Nutrient values: Logged meals, saved recipes, and photo scans are estimated by the AI model, not looked up in a food composition database.
• Barcode-scanned products: Nutrition for a scanned barcode comes from the Open Food Facts database, USDA branded-product data from USDA FoodData Central (https://fdc.nal.usda.gov), or the nutrition panel printed on the product, read directly rather than model-estimated.
• Provider imports: Meals imported from a connected provider (Cronometer, Apple Health, Health Connect, or Google Health) are recorded exactly as that provider reported them, not re-estimated by the app.
• Energy conversion: Calories are related to macronutrients using the Atwater general factors — 4 kcal/g for protein, 4 kcal/g for carbohydrate, 9 kcal/g for fat, and 7 kcal/g for alcohol. Reference: Merrill AL & Watt BK, "Energy Value of Foods: Basis and Derivation," USDA Agriculture Handbook No. 74, 1973.
• Daily targets and context: Calorie and macro targets shown alongside your log follow the USDA Dietary Guidelines for Americans (https://www.dietaryguidelines.gov) and the Institute of Medicine's Dietary Reference Intakes for macronutrients. Reference: Institute of Medicine, "Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids," National Academies Press, 2005.
• Portion estimation from photos and descriptions: AI-estimated portion sizes are approximations. Accuracy is inherently limited by what is visible or described; values carry a confidence indicator where applicable and are always editable.

TDEE (พลังงานที่ใช้ทั้งหมดต่อวัน)

The app estimates daily calorie burn using one of three methods, selected automatically based on available data:

• Mifflin-St Jeor equation (primary): BMR derived from weight, height, age, and sex, scaled by an activity multiplier from step count and logged workout intensity. Reference: Mifflin MD et al., "A new predictive equation for resting energy expenditure in healthy individuals," American Journal of Clinical Nutrition, 1990.
• Katch-McArdle equation (lean-mass path): Used when recent body composition data is available. BMR derived from lean mass rather than total weight. Reference: McArdle WD, Katch FI, Katch VL, Exercise Physiology: Nutrition, Energy, and Human Performance.
• Adaptive TDEE (back-calculation): When sufficient logging history exists, TDEE is back-calculated from the relationship between observed caloric intake and smoothed weight trend using exponentially weighted moving average and regression analysis. Conceptual basis: Hall KD et al., "Quantification of the effect of energy imbalance on bodyweight," The Lancet, 2011; Thomas DM et al., "Time to correctly predict the amount of weight loss with dieting," Journal of the Academy of Nutrition and Dietetics, 2014.

NSI (ดัชนีความแข็งแรงแบบปรับมาตรฐาน)

NSI scores a strength training set relative to population standards for the user's bodyweight, age, and sex.

• 1RM estimation: Derived from validated predictive equations using lifted weight and repetitions performed, with repetitions capped at 20 because higher-rep sets reflect muscular endurance rather than maximal strength. A single-repetition set is taken as measured rather than predicted, ตั้งแต่ the load lifted is itself the one-repetition maximum. Default equation: Epley. Reference: Epley B, "Poundage Chart," Boyd Epley Workout, 1985. Selectable alternatives include Lander and Brzycki. References: Lander J, "Maximums based on reps," National Strength and Conditioning Association Journal, 1985; Brzycki M, "Strength testing: predicting a one-rep max from reps-to-fatigue," Journal of Physical Education, Recreation & Dance, 1993.
• Population strength standards: Per-exercise benchmarks interpolated by bodyweight and adjusted by age factor, derived from published normative strength data and allometric scaling research. References: Haff GG & Triplett NT (eds.), Essentials of Strength Training and Conditioning, 4th ed., NSCA; Jaric S, "Muscle strength testing: use of normalisation for body size," Sports Medicine, 2002.
• Age adjustment: Strength standards are scaled by age to reflect the documented decline in peak force production across the lifespan. Reference: Pearson SJ et al., "Muscle function and ageing," Scandinavian Journal of Medicine & Science in Sports, 2002.

Fit Score (คะแนนรวมรายวัน 0-100)

The daily Fit Score combines training performance, sleep, nutrition, recovery, activity, and subjective wellbeing into a single 0-100 wellness index. Components re-normalize when data is missing for a given day. Underlying methodologies for each domain are documented in the individual metric sections on this page.

คะแนนการนอนหลับ (0-100)

When stage data is available from a connected wearable, the คะแนน uses four components. When only duration is available, the duration component scales to 100%.

• Duration: Scored against recommended nightly sleep duration for adults. Reference: Hirshkowitz M et al., "National Sleep Foundation's sleep time duration recommendations," Sleep Health, 2015.
• Deep sleep: Scored against published normative deep sleep values for adults. Reference: Ohayon M et al., "Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals," Sleep, 2004.
• REM sleep: Scored against published normative REM values for adults. Reference: Ohayon M et al., "Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals," Sleep, 2004; Carskadon MA & Dement WC in Principles and Practice of Sleep Medicine.
• Awakenings: Scored based on clinical sleep continuity guidelines. Reference: American Academy of Sleep Medicine, International Classification of Sleep Disorders.

คะแนนการฟื้นตัว (0-100)

Recovery Score combines physiological and subjective signals into a single readiness index.

• HRV: RMSSD values compared to the user's personal rolling baseline. Because HRV is highly individual, personal baseline is weighted more heavily than population norms. Reference: Plews DJ et al., "Training adaptation and heart rate variability in elite endurance athletes," Sports Medicine, 2013; Shaffer F & Ginsberg JP, "An overview of heart rate variability metrics and norms," Frontiers in Public Health, 2017.
• Resting heart rate: Compared to the user's personal baseline and population health norms. Lower RHR correlates with greater cardiovascular fitness and recovery. Reference: American Heart Association guidelines on resting heart rate as a cardiovascular risk marker.
• Sleep: Hours logged from wearable or manual entry. Reference: AASM 7-9 hour recommendation for adults.
• Wellbeing: Self-reported energy, mood, soreness, and stress. Soreness and stress are inverted (higher soreness/stress = lower recovery).

ความแปรปรวนของอัตราการเต้นของหัวใจ (HRV)

The app records and displays RMSSD (root mean square of successive RR interval differences) in milliseconds, the most validated short-term HRV metric for athletic monitoring.

• Measurement: Sourced from Apple Health, Fitbit, Oura, or Google Health Connect. Each provider uses their own validated capture method (overnight average or morning reading).
• Interpretation: Compared to the user's own rolling baseline, not population averages, because absolute HRV values vary dramatically between individuals. Reference: Plews DJ et al., "Comparison of heart-rate-variability recording with smartphone photoplethysmography, Polar H7 chest strap, and electrocardiography," International Journal of Sports Physiology and Performance, 2014.
• Readiness signal: A meaningful drop below personal baseline (especially combined with elevated RHR) is used as a recovery flag. Reference: Buchheit M, "Monitoring training status with HR measures: do all roads lead to Rome?" Frontiers in Physiology, 2014.