Iba 2017 SIC (Table 3; max 6). platelets_10e9_l >=150=0, <150 and >=100=1, <100=2. inr <=1.2=0, >1.2 and <=1.4=1, >1.4=2. sofa_four_items is the caller-assigned respiratory+cardiovascular+hepatic+renal SOFA sum (0-16; not coagulation, not CNS): 0=0, 1=1, >=2=2 (cap 2). sic true iff score >=4 and platelet_pts+inr_pts >=3. Not a sepsis or DIC diagnosis. Not ISTH overt DIC. Not JAAM-DIC. magent does not start anticoagulants.

Cucchiara 2008 sICH after rtPA for ischemic stroke. Four caller-assigned 1-point flags: age >60, baseline NIHSS >10, baseline glucose >150 mg/dL (8.325 mmol/L), platelets <150,000/mm3. Total 0-4. Any ICH 2.6/9.7/15.1/37.9% for scores 0/1/2/>=3 (scores 3 and 4 share 37.9). magent does not assay glucose or platelets or compute NIHSS. Not a tPA order and not a hemorrhage diagnosis.

Nadler 1962 adult blood volume plus exponential remaining-fraction washout (Y/Y0 = e^(−X)) for sickle RBC exchange replacement pRBC volume. FCR = target_hbs / initial_hbs; X = −ln(FCR). Caller supplies hematocrits and HbS percents. Optional approx units if unit_volume_ml is sent. Deterministic published arithmetic. Not an apheresis order or medical device.

Simon Broome Register 1991 adult definite familial hypercholesterolaemia: total cholesterol >7.5 mmol/L or LDL cholesterol >4.9 mmol/L plus tendon xanthomas in the patient or a relative (parent, child, grandparent, sibling, uncle, or aunt). One lipid (mmol/L or mg/dL). magent does not assay lipids or examine tendons. Not a diagnosis, not possible FH, and not a DNA or child-cutoff classifier.

Hennes 2008 simplified AIH (Table 2): caller-assigned autoantibody points 0-2 (max 2 across ANA/SMA/LKM/SLA), IgG points 0-2 (≤ULN / >ULN / >1.10×ULN), histology none 0 / compatible 1 / typical 2, viral hepatitis absent true 2. Max 8. Score <6 not_aih, 6 probable, ≥7 definite. Not a diagnosis and not the original 1999 IAHG score. magent does not titer autoantibodies, assay IgG vs ULN, or read liver histology.

Laughon 2011 simplified Bishop score from three caller-assigned Bishop 1964 integers: dilation, station, and effacement (each 0-3; max 9). Paper bins: dilation 0 / 1-2 / 3-4 / 5-6 cm; station -3 / -2 / -1 or 0 / +1 or +2; effacement 0-30 / 40-50 / 60-70 / 80%. favorable iff score >5 (Laughon compared with original Bishop >8; PPV 87.7% in that paper). magent does not examine the cervix or add consistency or position. Not a cervical exam or induction order.

Fisher 2010 SINS. Location junctional 3 / mobile 2 / semirigid 1 / rigid 0. Pain mechanical 3 / occasional_not_mechanical 1 / pain_free 0. Bone lytic 2 / mixed 1 / blastic 0. Alignment subluxation_translation 4 / de_novo_deformity 2 / normal 0. Collapse gt_50 3 / le_50 2 / no_collapse_gt_50pct_body 1 / none 0. Posterolateral bilateral 3 / unilateral 1 / none 0. Max 18. Integer sum only; not later SOSG strata, not TLICS, not an operative order. magent does not read imaging. Not a medical device.

Acker 2015 SIPA: heart rate in bpm divided by cuff systolic pressure in mm Hg, rounded to two decimals. Elevated if the ratio exceeds the age-band cutoff: 1.22 (ages 4-6), 1.0 (7-12), or 0.9 (13-16). Ages 4-16 only; not for infants or adults 17 and older. Not a diagnosis of shock and not the adult shock-index greater-than-0.9 rule. magent does not measure vitals.

SIRS from Bone 1992 ACCP/SCCM: four binary criteria (temperature >38 or <36 °C, heart rate >90, respiratory rate >20 or PaCO2 <32 mm Hg, WBC >12000 or <4000 per µL or bands >10%); present when ≥2 of 4. Deterministic published criteria with citation; not a diagnosis, not Sepsis-3, not qSOFA or SOFA.

Gladman 2002 SLEDAI-2K: weighted sum of 24 caller-assigned descriptors present in the last 10 days (max 105). magent does not examine the patient. Proteinuria scores if >0.5 g/24 h (not original SLEDAI >3.5 g/24 h only). Rash, alopecia, and mucosal ulcers score if new or persistent (not new only). Not a diagnosis, not SLICC damage, not SELENA-SLEDAI flare rules, and not a medical device.

SLICC 2012 SLE classification (Petri Table 3). classified iff biopsy-proven lupus nephritis plus ANA or anti-dsDNA, or at least four criteria including at least one clinical and one immunologic. 11 clinical and 6 immunologic caller flags. Direct Coombs counts only in the absence of hemolytic anemia. Criteria need not be concurrent. Not a diagnosis. Not 1997 ACR. Not 2019 EULAR/ACR. magent does not examine skin, assay serologies, or read a biopsy.

Vaccaro 2007 SLICS: morphology none 0 / compression 1 / burst 2 / distraction 3 / rotation_translation 4 (single most severe) + dlc intact 0 / indeterminate 1 / disrupted 2 + neuro intact 0 / root 1 / complete 2 / incomplete 3 + continuous_cord_compression +1 only with a deficit. Max 10. Bands 0-3 / 4 / >=5 treatment guidance. Not a surgical order. Not TLICS (thoracolumbar). Not AO spine. Not SINS. magent does not read imaging. Not a medical device.

Charles 2008 SMART-COP for intensive respiratory or vasopressor support in community-acquired pneumonia: SBP <90 (2), multilobar CXR (1), albumin <3.5 g/dL (1), age-adjusted RR (1), HR >=125 (1), new confusion (1), age-adjusted low oxygen (2), arterial pH <7.35 (2; omitted pH scores 0). Max 11. IRVS 0-2 low, 3-4 moderate, 5-6 high, >=7 very high. Not a site-of-care protocol. Albumin g/dL or g/L. Exactly one of PaO2, SpO2, or P/F.

Selzer 1975 13-item Short Michigan Alcoholism Screening Test (SMAST). Caller-assigned yes/no items; score is the count of answers in the alcoholic direction using that paper's key (feel_normal_drinker, friends_think_normal_drinker, and able_to_stop score 1 if false; the other ten score 1 if true), max 13. Bands 0-1 nonalcoholic, 2 possible, ≥3 alcoholic. Not a diagnosis, not the 25-item MAST, not Pokorny 1972 Brief MAST, and not SMAST-G. magent does not interview the patient.

Lakshman 2018 2/20/20 smoldering myeloma risk: one point each for BMPC >20%, serum M-protein >2 g/dL, and involved/uninvolved FLC ratio >20 (strict greater-than). 0 factors low (median TTP 110 months), 1 intermediate (68), 2 or 3 high (29). Not Mayo 70/20, not PETHEMA, not myeloma ISS, not a treatment order. BMPC 10-59 and FLC 0.01-99.9 (IMWG 2014 SLiM myeloma is rejected). magent does not assay marrow, M-protein, or free light chains.

Gill 2005 Simplified Motor Score (SMS) from one caller-assigned motor category: obeys (2, obeys commands), localizes (1, localizes pain), or withdrawal_or_less (0, withdrawal to pain or less). Companion to GCS, not a replacement protocol and not a triage or intubation order. magent does not examine the patient. Deterministic published score with citation.

Sodium deficit from weight, sex, and measured Na using Adrogué–Madias TBW (0.6 male / 0.5 female). Deterministic published arithmetic with citation; not a sodium infusion protocol.

SOFA (Sepsis-related Organ Failure Assessment) from Vincent 1996 Table 3: six organs 0-4, max 24, from PaO2/FiO2 with a ventilation gate, platelets, bilirubin, MAP and adrenergic agents, caller-assigned GCS, and creatinine or urine output. Deterministic published score with citation; not a Sepsis-3 diagnosis.

Ranzani 2025 SOFA-2 from six organs 0-4 (brain, respiratory, cardiovascular, liver, kidney, hemostasis), max 24. Maps PaO2/FiO2 or SpO2/FiO2 with NIV/ECMO, MAP and vasopressor dose bands, caller-assigned GCS and delirium-drug flag, bilirubin, platelets, creatinine or urine rate, and RRT. Not Vincent 1996 SOFA and not a diagnosis.

Sokal 1984 relative-risk index for newly diagnosed untreated chronic granulocytic leukemia (CML). sokal_index = exp(0.0116*(age-43.4)+0.0345*(spleen_cm-7.51)+0.188*((platelets_10e9_l/700)^2-0.563)+0.0887*(blasts_percent-2.10)). Low <0.8, intermediate 0.8–1.2, high >1.2. Coefficient 0.188. Does not diagnose CML or choose TKI therapy.

sPESI (simplified Pulmonary Embolism Severity Index) from Jiménez 2010: 1 point each for age >80, cancer, chronic cardiopulmonary disease (one combined item), heart rate ≥110, systolic BP <100 mm Hg, and SpO2 <90%; max 6. Score 0 is low risk; ≥1 is high risk. Only for already diagnosed acute PE as in the paper. Not a diagnosis, not an outpatient-treatment protocol, and not original PESI classes.

Frank 2002 SSIGN (stage, size, grade, necrosis) for clear cell RCC after radical nephrectomy. 1997 TNM: pT1 0, pT2 1, pT3 2, pT4 0; N0/Nx 0, N1/N2 2; M0 0, M1 4; size ≥5 cm 2; nuclear grade 1–2 0, 3 1, 4 3; coagulative necrosis 2; max 15. Returns the published CSS row. Not 2002/2010 TNM. magent does not stage or read pathology. Not a diagnosis. Not a medical device.

Hoddes 1973 Stanford Sleepiness Scale degree 1-7 as assigned by the caller. Returns the degree, score (same as the degree), max 7, and published wording (feeling active and vital; alert; wide awake through almost in reverie; sleep onset soon). Not the Epworth Sleepiness Scale. Not a sleep-disorder diagnosis; magent does not observe the patient.

Steinhart 2009 Equation 1 probability of acute heart failure from caller-assigned age, clinician pretest probability (fraction 0-1 or 0-100 percent), and NT-proBNP in pg/mL. Returns estimated_probability_percent from the unrounded probability plus published ≤0.20 rule-out and ≥0.80 rule-in flags. Not a diagnosis, diuretic, or admission order. magent does not assay NT-proBNP.

Gomez 2016 Step-by-Step for febrile infants 90 days and younger (derivation Mintegi 2014). High if ill-appearing, age 21 days or younger, leukocyturia, or PCT >=0.5 ng/mL; else intermediate if CRP >20 mg/L or ANC >10000/uL; else low. Returns risk, low_risk, high_risk, and intermediate_risk. magent does not examine the infant or assay UA, PCT, CRP, or ANC. Not Rochester, Boston, Philadelphia, or PECARN febrile. Not a medical device and not a discharge, admission, LP, or antibiotic order.

Convert a glucocorticoid milligram dose among hydrocortisone, prednisone, and dexamethasone using Meikle 1977 equivalent doses (20 : 5 : 0.75). Not a taper, stress-dose protocol, or regimen recommendation; mineralocorticoid activity is not converted. Deterministic published table with citation.

Okhunov 2013 STONE nephrolithometry for kidney calculi (PCNL). Caller-assigned published item points: size 0_399 1 / 400_799 2 / 800_1599 3 / ge_1600 4 (mm²); tract le_100 1 / gt_100 2 (skin-to-stone mm); obstruction none_or_mild 1 / moderate_or_severe 2 (hydronephrosis); calyces 1_or_2 1 / 3 2 / staghorn 3; essence le_950 1 / gt_950 2 (HU). Sum 5-13. No published SFR bands. Not the Moore 2014 ureteral STONE score. magent does not measure CT. Not a medical device. Not a PCNL order.

Moore 2014 STONE score for ED prediction of uncomplicated ureteral stone. Table 3 integer points: sex female 0 male 2; timing gt_24h 0, 6_to_24h 1, lt_6h 3; origin black 0, nonblack 3 (caller-assigned as used in Moore 2014; magent does not assign race); nausea none 0, nausea_alone 1, vomiting 2; hematuria absent 0, present 3. Sum 0-13; 0-5 low, 6-9 moderate, 10-13 high. Not a CT order or stone diagnosis. Not S.T.O.N.E. nephrolithometry.

Eherer–Fordtran 1992 fecal osmotic gap: stool osmolality minus 2×(stool Na + stool K). Assumed stool osm 290 mOsm/kg when measured osm is omitted. Deterministic published arithmetic with citation; not a diagnosis of osmotic versus secretory diarrhea.

Chung 2008 Appendix 2 STOP-Bang OSA screen. Four caller-assigned STOP flags (loud snoring heard through closed doors, daytime tiredness, observed stop-breathing, hypertension) plus computed BANG: BMI >35, age >50, neck >40 cm, male; each 1; max 8. BMI 35, age 50, and neck 40 cm do not score. Returns bands 0-2 and >=3. Not Chung 2012 three-tier, not a medical device, and not a sleep-study order.

Battle 2014 STUMBL integer score after blunt chest wall trauma: 1 point per decade of age starting at 10, 3 per rib fracture, 5 for chronic lung disease, 4 for preinjury anticoagulants, and 2 per 5% SpO2 decrease starting at 94%. Table 5 complication probabilities 13/29/52/70/80/88%. magent does not examine the chest or count ribs. Not a medical device, not a disposition or ICU order.

Driver 2017 4-variable formula for normal-variant STE in V2-V4 (early repolarization) versus subtle LAD occlusion: formula_value = 0.052*QTc_ms - 0.151*QRS_V2_mm - 0.268*R_V4_mm + 1.062*STE60_V3_mm, 3 decimals. lad_occlusion_suggested is true iff the value is ≥18.2 (derivation highest-accuracy cutpoint). Intended only when ≥1 mm STE in V2-V4 without obvious occlusion features. magent does not read the ECG. Not a STEMI diagnosis, cath-lab order, or 3-variable formula. Not a medical device.

Ohle 2025 Sudbury vertigo ED score (Table 2): male 1, age>65 flag 1, diabetes 1, hypertension 3, motor or sensory deficit 5, cerebellar signs 6, BPPV -5. Returns score (female+BPPV -5, not clamped), max 17, Table 3 bands lt_5 / 5_to_8 / gt_8 (0.0 / 2.1 / 41.0%), and investigate_further if score >4. Not a medical device, stroke diagnosis, CT order, HINTS, or HINTS-plus. magent does not examine the patient. Age is the age>65 flag, not years.

Waldron 2023 SULF-FAST for patients who already report a TMP-SMX (sulfa antibiotic) allergy. Adapted PEN-FAST flags, not PEN-FAST itself. Three caller-assigned flags: allergy event 5 or fewer years ago (2), anaphylaxis/angioedema or potential SCAR as one criterion (2), treatment required (1). Max 5. Risk: low if score < 3, elevated if ≥ 3. low_risk_for_testing iff score < 3. Not an allergy diagnosis or delabeling order. magent does not take an allergy history.

Intraoperative Surgical Apgar (Gawande 2007 Table 1) from estimated blood loss, lowest mean arterial pressure, and lowest heart rate, summed to a 0–10 total. Returns score_le_4 because scores ≤4 were associated with 58.6% 30-day death or major complication in the validation cohort (9–10 associated with 3.6%). Deterministic published arithmetic with citation; not Virginia Apgar; not operative clearance.

Strander 2005 Swede score: five caller-assigned 0-2 items (acetowhiteness, margins_surface, vessels, lesion_size, iodine; Table I final rounded weights; max 10). swede_band <5, 5-7, or >=8 (≥5 identified all HGL in that derivation set; ≥8 had 90% specificity). magent does not read the colposcope. Not a CIN diagnosis, see-and-treat order, or Reid index.

Body 2016 T-MACS logistic ACS probability from caller-assigned ECG ischaemia, worsening angina, right-arm radiation, vomiting, observed sweating, hypotension (SBP <100), and Roche Elecsys hs-cTnT ng/L. Uses that paper's rounded coefficients. Bands p<0.02 very_low, 0.02<=p<0.05 low, 0.05<=p<0.95 moderate, p>=0.95 high. magent does not interpret ECG or assay troponin. Not original MACS, HEART, EDACS, or Vancouver. Not a diagnosis or medical device.

Yücel 2006 TASH (Trauma Associated Severe Hemorrhage): seven independent nested less-than point bands for arrival SBP, hemoglobin, intra-abdominal fluid, complex long-bone/pelvic fracture AIS, heart rate, base excess, and male sex; max 28. Highest matching band only (do not stack). Deterministic published arithmetic with citation; not a transfusion or MTP order; not ABC (assessment of blood consumption). Caller assigns FAST and AIS; magent does not read imaging.

Terrapon 2021 Therapy-Disability-Neurology (TDN) grade. Caller assigns therapy t1-t5, modified Rankin 0-6 (6 is death), and new neurologic deficit. Overall grade is max(T, D, N); t5 or mRS 6 is always 5. D maps mRS 0-1→1, 2-3→2, 4→3, 5→4, 6→5; new deficit is N=2 else N=1. Not a diagnosis and not Clavien-Dindo alone. magent does not grade the event.

Thakar 2005 clinical score for acute renal failure requiring dialysis after open-heart surgery. Points from sex, CHF, LVEF <35%, preoperative IABP, COPD, insulin-requiring diabetes, previous cardiac surgery, emergency surgery, procedure type (CABG / valve / CABG+valve / other), and preoperative creatinine (<1.2 / 1.2 to <2.1 / >=2.1 mg/dL). Returns integer score (max 17) and category 0–2 / 3–5 / 6–8 / 9–13. Not a dialysis order; not AKIN or RIFLE staging.

THRIVE (Totaled Health Risks in Vascular Events; Flint 2010 AJNR) from trichotomized age (<=59 0, 60-79 1, >=80 2), trichotomized NIHSS (<=10 0, 11-20 2, >=21 4), and chronic disease scale (hypertension, diabetes, atrial fibrillation; 1 each). Score 0-9; bands low 0-2, moderate 3-5, high 6-9 with published 90-day mRS 0-2 and mortality percents. Not a treatment recommendation. magent does not examine the patient or assign NIHSS.

Morrow InTIME-II TIMI risk index: heart rate × (age/10)^2 / systolic BP, with the five published groups (≤12.5, >12.5–17.5, >17.5–22.5, >22.5–30, >30). Deterministic published arithmetic with citation; not a reperfusion decision.

TIMI risk score for ST-elevation MI from age, diabetes/hypertension/angina, systolic BP, heart rate, Killip class, weight, anterior STE or LBBB, and time to treatment. Returns a 0-14 point total with components. Deterministic published score with citation; not a reperfusion or fibrinolytic protocol.

TIMI risk score for unstable angina / NSTEMI from age and six caller-assigned flags. Returns the 0-7 point total and components. Deterministic published score with citation; not an invasive-strategy protocol.

Tisdale 2013 QT prolongation risk score (Table 5): age ≥68 (1), female (1), loop diuretic (1), potassium ≤3.5 mmol/L (2), admission QTc ≥450 ms (2), acute MI (2), sepsis (3), heart failure (3), one QTc-prolonging drug (3) plus ≥2 drugs (3 stacked; max 21). Returns bands 0-6 / 7-10 / ≥11. Deterministic published arithmetic with citation; not a medical device, not a drug-stop protocol, and not a TdP diagnosis.

Vaccaro 2005 TLICS: morphology compression 1 / burst 2 / translational_rotational 3 / distraction 4 (single most severe) + plc intact 0 / suspected_indeterminate 2 / injured 3 + neuro intact 0 / nerve_root 2 / complete_cord_conus 2 / incomplete_cord_conus 3 / cauda_equina 3. Max 10. Bands 0-3 nonoperative, 4 equivocal, >=5 operative guidance. Not a surgical order. Not AO spine. Not SINS (neoplastic). magent does not read imaging. Not a medical device.

Revised Tokuhashi 2005 (0-15). KPS poor/moderate/good 0/1/2. Extraspinal ge_3/one_or_two/none 0/1/2. Vertebral ge_3/two/one 0/1/2. Visceral unremovable/removable/none 0/1/2. Primary lung osteosarcoma stomach bladder esophagus pancreas 0; liver gallbladder unidentified 1; others 2; kidney uterus 3; rectum 4; thyroid prostate breast carcinoid 5. Palsy complete/incomplete/none 0/1/2. Bands 0-8/9-11/12-15. Not an operative order. Not 1990. Not SINS. Does not read imaging. Not a medical device.

Kiriyama 2018 TG18 (=TG13) cholangitis. Definite if one A (fever/chills or elevated WBC/CRP), one B (jaundice or abnormal LFTs), and one C (dilatation or etiology); suspected if A plus B or C; else none. Grade III if any organ dysfunction, else Grade II if any two of WBC, fever >=39C, age >=75, bilirubin >=5, or hypoalbuminemia, else Grade I. Severity none if diagnosis none. Not cholecystitis or Alvarado/RIPASA. magent does not examine, assay, or read imaging. Not a medical device.

Yokoe 2018 TG18 (=TG13) acute cholecystitis. Definite if local signs (Murphy or RUQ mass/pain/tenderness), systemic signs (fever, elevated CRP or WBC), and imaging; suspected if local and systemic without imaging; else none. Suspected/definite: Grade III if any organ dysfunction, else Grade II if any Grade II flag, else Grade I. Severity none when diagnosis is none. Not cholangitis or Alvarado/RIPASA. magent does not examine, assay, or read imaging. Not a medical device.

Kaplan 1990 estimated plasma volume for therapeutic plasma exchange: epv_l = 0.065 * weight_kg * (1 - hematocrit_percent/100). Returns one plasma volume in liters and milliliters. Hematocrit is a percent (10-70), not a 0-1 fraction. Deterministic published arithmetic with citation. Not an apheresis order, Nadler blood volume, Ganzoni iron deficit, or a 1.5-PV / removal-percent table.