A point prescription is easier to review when the note shows the reasoning between pattern differentiation, treatment principle (treatment principle), and selected points (point selection). The purpose is not to create one universal formula, but to make the practitioner’s clinical logic visible.
For licensed-practitioner education. This article is not patient-specific medical advice and does not recommend diagnosis or treatment.
Begin with the working assessment
Point selection follows the current presentation, not an isolated symptom label. The Assessment should identify the working pattern and the findings that support it. A point that appears in many reference cases may still be irrelevant to the person in front of the practitioner.
A reviewable reasoning chain begins with explicit evidence. If the pattern is uncertain, the proposed point strategy should preserve that uncertainty rather than overstate precision.
State the treatment principle before listing points
The treatment principle translates the assessment into an intended clinical direction. It gives the reader a way to understand why the point combination was assembled. Without it, the point list can look like a memorized formula disconnected from the visit.
The principle should be concise and specific enough to guide the plan. It may address root and branch (root and branch), identify what is primary for the visit, and clarify whether the practitioner is continuing or revising an earlier strategy.
Describe roles within the point combination
A point prescription can be documented as more than a list. Grouping points by role—core strategy, channel relationship, local support, symptom-focused addition, or balancing pair—can make the formula easier to inspect. The exact taxonomy varies by practitioner and tradition.
Documentation does not need a paragraph for every point. A short explanation for the main combination, plus any non-obvious additions or substitutions, often captures the clinical intent.
Use channel and point relationships carefully
Channel or meridian (channel system) relationships, point categories, anatomical location, and practitioner lineage may all influence selection. Software should not collapse these into a single rule. A suggestion engine needs to surface the basis of a proposed relationship and allow the practitioner to reject it.
Terminology also matters. “Meridian” is a common English rendering, but the literal term for longitude is not the correct clinical term for this context.
What a citation can and cannot show
A retrieved source excerpt can show that a pattern, principle, or point relationship appeared in a relevant de-identified case. It can help a practitioner compare reasoning and find precedents that would otherwise be difficult to search.
A citation cannot establish that the same plan is appropriate now. Case similarity is not equivalence, and historical frequency is not a substitute for assessment. Citations support clinical review; they do not authorize care.
Designing AI point support responsibly
A responsible interface shows the proposed pattern, treatment principle, point suggestions, and source rationale as separate, editable layers. It identifies conflicting findings, avoids hiding source gaps, and records practitioner changes.
ASKLEMER is being built toward this cited, practitioner-in-the-loop workflow. It will suggest and assist; the licensed practitioner remains responsible for point selection and the final plan.
Key takeaway
Clinical AI is most useful when it keeps facts, inferences, sources, and practitioner decisions distinct. ASKLEMER is in development and makes no claim of clinical performance or availability.