Modeling and Decision Application of Pregnancy Medication Cognition Path Based on DIKWP
Yucong Duan
Benefactor: Zhendong Guo
International Standardization Committee of Networked DIKWP for Artificial Intelligence Evaluation(DIKWP-SC)
World Artificial Consciousness CIC(WAC)
World Conference on Artificial Consciousness(WCAC)
(Email: duanyucong@hotmail.com)
Overview
Clinical decision-making regarding medication use during pregnancy is highly challenging: nearly all pregnant women may require medication, yet reliable data on safety are extremely limited (Medicine and Pregnancy: An Overview | Medicine and Pregnancy | CDC). During pregnancy, many drugs lack sufficient clinical trial evidence, leaving healthcare providers and patients to make decisions with incomplete information and high uncertainty (Medicine and Pregnancy: An Overview | Medicine and Pregnancy | CDC). On the other hand, certain untreated conditions pose greater risks to both mother and fetus (Medicine and Pregnancy: An Overview | Medicine and Pregnancy | CDC)—for example, uncontrolled hypertension, diabetes, or hypothyroidism can lead to severe consequences such as miscarriage or fetal growth restriction (Levothyroxine: a medicine for an underactive thyroid (hypothyroidism) - NHS). Therefore, medication decisions during pregnancy require balancing data uncertainty, information-layer interference, knowledge limitations, and multiple objectives.
Professor Yucong Duan’s DIKWP model offers a novel approach to such complex decision-making. DIKWP represents five cognitive levels: Data (D), Information (I), Knowledge (K), Wisdom (W), and Purpose (P) (Based on the DIKWP Network Model 3 – Research Notes). Unlike the traditional linear DIKW pyramid, the DIKWP model treats these five elements as an interactive network, where any two levels can transform bidirectionally, forming a 5×5 matrix of 25 cognitive pathways (Based on the DIKWP Network Model 3 – Research Notes). This networked DIKWP cognitive model features dynamic multidimensionality, allowing outputs to feed back into inputs, thereby creating a closed-loop cognitive system (Based on the DIKWP Network Model 3 – Research Notes). Under the open-world assumption, the DIKWP model addresses cognitive challenges through "intent-driven" methods: even when faced with incomplete, inconsistent, and imprecise information (referred to as the "3-No Problem" by Yucong Duan (Based on the DIKWP Network Model 3 – Research Notes)), the system can dynamically generate semantic associations guided by the Purpose layer (P), optimize decisions at the Wisdom layer (W), and construct solutions that meet multiple objectives (Based on the DIKWP Network Model 3 – Research Notes).
In summary, the DIKWP model holds unique value for clinical decision-making during pregnancy. First, it addresses data uncertainty through interactions between the Data and Information layers, integrating fragmented test results, case reports, and other data into meaningful information (Based on the DIKWP Network Model 3 – Research Notes). Second, when confronted with conflicting or inconsistent information from various sources, the model leverages feedback from the Wisdom layer to reconcile discrepancies and reconstruct logic at the Knowledge layer (Based on the DIKWP Network Model 3 – Research Notes). Third, for knowledge gaps due to the lack of large-scale trials in pregnancy, the model adopts an intent-driven approach to infer and extrapolate possible knowledge within the semantic space (e.g., extrapolating missing medication experiences based on existing guidelines) (Based on the DIKWP Network Model 3 – Research Notes). Finally, to balance the multiple objectives of maternal and fetal health, the DIKWP model introduces shared goals of both physicians and patients (e.g., alleviating maternal symptoms while ensuring fetal safety) at the Purpose layer, ensuring these goals permeate the entire decision-making chain from Data → Information → Knowledge → Wisdom (Based on the DIKWP Network Model 3 – Research Notes).
For these reasons, the DIKWP model is particularly well-suited for complex scenarios like medication use during pregnancy. When faced with high uncertainty and multi-objective trade-offs, it provides a clear cognitive pathway framework and robust decision support.
Text Table Analysis (DIKWP 25-Path Scoring)
Below, we analyze several common medication scenarios during pregnancy, comparing two representative drugs in each scenario using the DIKWP model. Based on the original networked DIKWP model proposed by Professor Yucong Duan, we assign relative scores (1–5, where 5 indicates the highest degree of participation or importance) to each drug across the 25 cognitive transformation paths. The accompanying text briefly explains the differences in cognitive processing between the two drugs at each level.
Common Medication Scenario: Antipyretic and Analgesic Use
Drugs Compared: Acetaminophen vs. Ibuprofen
Cognitive Transformation Path |
Acetaminophen Score |
Ibuprofen Score |
D→D |
3 |
3 |
D→I |
3 |
4 |
D→K |
3 |
4 |
D→W |
5 |
2 |
D→P |
2 |
4 |
I→D |
1 |
3 |
I→I |
2 |
4 |
I→K |
3 |
5 |
I→W |
2 |
4 |
I→P |
2 |
5 |
K→D |
1 |
4 |
K→I |
2 |
5 |
K→K |
3 |
4 |
K→W |
5 |
4 |
K→P |
2 |
5 |
W→D |
1 |
4 |
W→I |
2 |
4 |
W→K |
2 |
4 |
W→W |
3 |
3 |
W→P |
2 |
4 |
P→D |
2 |
5 |
P→I |
2 |
5 |
P→K |
3 |
5 |
P→W |
1 |
5 |
P→P |
1 |
5 |
Explanation:
In the context of antipyretic and analgesic use during pregnancy, the cognitive pathway distributions of acetaminophen (paracetamol, e.g., Tylenol) and ibuprofen differ significantly.
·Acetaminophen, as the first-line antipyretic and analgesic in pregnancy, relies more on rapid-response pathways from Data directly to Wisdom (D→W, score 5). Clinically, when a pregnant woman presents with fever or pain, physicians can confidently prescribe acetaminophen with minimal hesitation, as extensive historical data and knowledge support its safety and efficacy (Nonsteroidal Anti-Inflammatory Drugs (NSAIDs): Drug Safety Communication - Avoid Use of NSAIDs in Pregnancy at 20 Weeks or Later | FDA).
·In contrast, ibuprofen triggers more purpose-driven, cautious deliberation pathways (P→W, score 5). Physicians must carefully weigh its intended benefits against potential risks (e.g., fetal renal impairment and oligohydramnios if used after 20 weeks of gestation (Nonsteroidal Anti-Inflammatory Drugs (NSAIDs): Drug Safety Communication - Avoid Use of NSAIDs in Pregnancy at 20 Weeks or Later | FDA)). Thus, ibuprofen requires significantly higher engagement at the Wisdom (W) and Purpose (P) levels, reflecting the need for advanced judgment to justify its use (Application of the DIKWP Model in Drug Selection: Acetaminophen vs. Ibuprofen - Zhihu Column).
·At the Knowledge (K) level, clinical guidelines and pharmacopeias consistently recommend acetaminophen as the preferred choice, while cautioning against NSAIDs like ibuprofen (Nonsteroidal Anti-Inflammatory Drugs (NSAIDs): Drug Safety Communication - Avoid Use of NSAIDs in Pregnancy at 20 Weeks or Later | FDA). This difference is evident in the K→P scores (acetaminophen: 2; ibuprofen: 5), indicating that ibuprofen use must be strongly purpose-driven (e.g., reserved for cases where alternatives fail and only during specific safer periods in mid-pregnancy).
·Summary:
oAcetaminophen excels in low-level, direct pathways (Data/Information → Decision), requiring minimal cognitive effort due to its well-established safety profile.
oIbuprofen demands high-level, feedback-driven pathways (Purpose/Wisdom → Decision), necessitating careful risk-benefit analysis.
oThis aligns with clinical experience: acetaminophen is straightforward for single-purpose use, whereas ibuprofen requires cautious, context-dependent decision-making (Application of the DIKWP Model in Drug Selection: Acetaminophen vs. Ibuprofen - Zhihu Column).
Acid Suppression: Ranitidine vs. Omeprazole
Cognitive Transformation Path |
Ranitidine Score |
Omeprazole Score |
D→D |
3 |
2 |
D→I |
3 |
4 |
D→K |
3 |
4 |
D→W |
5 |
2 |
D→P |
2 |
3 |
I→D |
1 |
2 |
I→I |
2 |
3 |
I→K |
3 |
5 |
I→W |
2 |
4 |
I→P |
2 |
4 |
K→D |
1 |
3 |
K→I |
2 |
5 |
K→K |
3 |
4 |
K→W |
5 |
4 |
K→P |
2 |
4 |
W→D |
1 |
3 |
W→I |
2 |
4 |
W→K |
2 |
4 |
W→W |
3 |
3 |
W→P |
2 |
4 |
P→D |
2 |
4 |
P→I |
2 |
4 |
P→K |
3 |
5 |
P→W |
1 |
4 |
P→P |
1 |
4 |
Explanation:
Both ranitidine (an H₂-receptor antagonist) and omeprazole (a proton pump inhibitor, PPI) are used to relieve acid reflux and hyperacidity during pregnancy. The table reveals distinct cognitive pathway patterns:
·Ranitidine scores higher in direct decision-making pathways (e.g., D→W = 5), indicating that clinicians often prescribe it immediately upon observing symptoms like heartburn, given its well-documented safety and efficacy in pregnancy.
·Omeprazole, however, dominates in information- and knowledge-driven pathways (e.g., I→K = 5, P→K = 5). Its use requires careful integration of patient-specific data (e.g., symptom severity, poor response to H₂ blockers) and pharmacological knowledge (e.g., FDA pregnancy category B/C, animal study risks but no clear human teratogenicity) (Treatment of Gastroesophageal Reflux Disease During Pregnancy) (Treating gastro-oesophageal reflux disease during pregnancy and lactation: what are the safest therapy options? - PubMed).
·Purpose-driven deliberation (P→W = 4 for omeprazole) is more critical for PPIs—they are reserved for severe reflux cases where H₂ blockers fail, balancing maternal comfort and fetal safety (Prilosec vs. Zantac for Acid Reflux).
·Historically, guidelines favored ranitidine as first-line, but recent evidence supports omeprazole’s safety (Use of cimetidine, omeprazole, and ranitidine in pregnant ... - PubMed), reflected in their similar K→K scores (3 vs. 4).
·Summary:
oRanitidine excels in low-level, rapid-decision pathways due to its established safety.
oOmeprazole requires high-level cognitive engagement, justifying its use only after thorough risk-benefit analysis in refractory cases.
Laxatives: Lactulose vs. Cisapride
Cognitive Transformation Path |
Lactulose Score |
Cisapride Score |
D→D |
3 |
1 |
D→I |
4 |
3 |
D→K |
4 |
3 |
D→W |
5 |
1 |
D→P |
3 |
4 |
I→D |
2 |
3 |
I→I |
3 |
3 |
I→K |
4 |
5 |
I→W |
3 |
4 |
I→P |
2 |
5 |
K→D |
2 |
4 |
K→I |
3 |
5 |
K→K |
3 |