The jury found Stewart guilty, and the district court sentenced him to 132 months in prison based on a criminal-history score of two. Stewart appeals.
ISSUES
I. Is the record evidence sufficient to support Stewart's conviction?
II. Did the district court abuse its discretion by admitting Dr. Swenson's expert testimony about the cause of B.G.D.'s injuries?
III. Did the district court err in calculating Stewart's criminal-history score?
ANALYSIS
I. Sufficient evidence supports Stewart's conviction.
In reviewing a claim of insufficient evidence, we view the evidence in the light most favorable to the verdict "to determine whether the facts in the record and the legitimate inferences drawn from them would permit the jury to reasonably conclude that the defendant was guilty beyond a reasonable doubt of the offense of which he was convicted." State v. Hanson , 800 N.W.2d 618, 621 (Minn. 2011) (quotations omitted). Where, as here, the challenged conviction is based on circumstantial evidence, we apply a two-step analysis. State v. Harris , 895 N.W.2d 592, 598-601 (Minn. 2017). First, we identify the circumstances proved "by resolving all questions of fact in favor of the jury's verdict," in deference to the jury's credibility determinations. Id. at 600. Second, we independently consider the "reasonable inferences that can be drawn from the circumstances proved." Id. at 601. The circumstances proved must, when viewed as a whole, "be consistent with a reasonable inference that the accused is guilty and inconsistent with any rational hypothesis except that of guilt." Id. But we "will not overturn a guilty verdict on conjecture alone." State v. Hayes , 831 N.W.2d 546, 553 (Minn. 2013).
To convict Stewart of first-degree assault, the state was required to prove that he intentionally inflicted great bodily harm upon B.G.D. Minn. Stat. §§ 609.02, subd. 10(2), .221, subd. 1 (2014). Stewart does not *674dispute that B.G.D.'s severe brain and eye injuries constitute great bodily harm.3 He contends that insufficient evidence supports the jury's finding that he caused those injuries.
Stewart rests his challenge principally on a narrow view of the circumstances proved. He argues that Dr. Swenson's expert medical testimony cannot establish the circumstances proved because Dr. Swenson drew inferences from certain facts. This argument is unavailing. The two-step circumstantial-evidence analysis does not distinguish between observed facts (e.g., a dark spot on an MRI film) and inferred facts (e.g., the child has a subdural hemorrhage ), but between the observed or inferred facts that establish the circumstances of an alleged offense and the inference from those circumstances that the required elements of the offense exist. See Harris , 895 N.W.2d at 599 & n.4 (noting distinction between "a finding that an alleged fact (which does not by itself establish the required element) exists and ... a conclusion that if the alleged fact exists, one can reasonably infer that the required element also exists"). The supreme court has consistently treated the nature of injuries and their possible causes as medical "facts" that a reviewing court must take as proved. See Hayes , 831 N.W.2d at 553 (including in circumstances proved medical testimony that child victim's "cranial trauma was more severe than could be explained by a mere accidental household fall"); State v. Hokanson , 821 N.W.2d 340, 355 (Minn. 2012) (including in circumstances proved that child victim "did not have any blood disorder or other medical problem that would have caused him to bruise easily"); State v. Rhodes , 657 N.W.2d 823, 841 (Minn. 2003) (noting that medical testimony about the cause of a victim's injuries "supports the jury's verdict" and that the jury "apparently disregarded" contrary medical testimony). Thus, when a jury is presented with conflicting medical testimony about the nature of injuries and their possible causes, we assume the jury believed the expert testimony that is most consistent with its verdict.
Applying that standard here, viewing all of the evidence, including the experts' testimony, in the light most favorable to the verdict, we identify the following circumstances proved regarding the cause of B.G.D.'s injuries. B.G.D. exhibited all three of the significant markers of abusive head trauma-bleeding around the brain, injury to the brain itself, and retinal hemorrhages. Like approximately half of children with abusive head trauma, he did not exhibit additional physical injuries such as bruises or fractures. B.G.D. did not experience any severe accidental trauma, such as a high-speed motor-vehicle accident, that could explain his injuries, and could not have caused them by banging his head against the floor. Medical conditions that can cause bleeding around the brain are rare, and B.G.D.'s lab results and medical history indicated no underlying conditions that could explain his condition. Children with severe brain injuries generally present symptoms immediately and are unable to behave normally once the injuries occur; B.G.D. became symptomatic and was determined to have "massive" brain injuries immediately after being alone with Stewart.
*675Overall, these circumstances support a reasonable inference that Stewart intentionally inflicted great bodily harm upon B.G.D. Stewart acknowledges as much. We therefore consider whether other rational hypotheses of causation exist. In doing so, we first observe that, unlike many other cases of assault against young children, there was no claim or evidence that B.G.D. suffered an accidental trauma that would explain his injuries, e.g. , Hayes , 831 N.W.2d at 553-54, or that another adult caused them, e.g. , Hokanson , 821 N.W.2d at 355-56. The remaining hypotheses are few and implausible. It is unreasonable to infer that B.G.D. caused the injuries himself by banging his head on the floor because Dr. Swenson explained (and we assume the jury believed) that was physically impossible. And additional evidence confirms that explanation-the impact of banging his head on the floor left no external mark, it caused him no distress, and he had similarly banged his head before without incident. Nor is it reasonable to infer that the injuries were the result of a rare condition that was not revealed in B.G.D.'s medical history or extensive lab testing. Accordingly, the only reasonable inference from the circumstances proved is that Stewart intentionally inflicted great bodily harm upon B.G.D.
II. The district court did not abuse its discretion by denying Stewart's motion in limine and admitting Dr. Swenson's expert testimony about the cause of B.G.D.'s injuries.
"Rulings concerning the admission of expert testimony generally rest within the sound discretion of the district court and will not be reversed absent a clear abuse of discretion." State v. Mosley , 853 N.W.2d 789, 798-99 (Minn. 2014). A defendant claiming that the district court erred in admitting evidence must demonstrate both abuse of discretion and resulting prejudice. State v. Stewart , 643 N.W.2d 281, 292 (Minn. 2002).
Expert opinion testimony is generally admissible if the witness is qualified as an expert, the expert's testimony is helpful to the jury, and the expert's opinion has "foundational reliability." Minn. R. Evid. 704. Stewart does not dispute that Dr. Swenson is a qualified medical expert whose testimony about abusive head trauma was helpful to the jury.4 But he argues the district court abused its discretion by admitting her testimony that B.G.D.'s injuries were caused by abusive head trauma because (1) the state failed to demonstrate in response to his pretrial motion that the opinion had "foundational reliability" and (2) the opinion constituted improper testimony as to the element of intent. We address each argument in turn.
Foundational Reliability
The proponent of expert opinion testimony bears the burden of showing that the testimony has foundational reliability. Doe v. Archdiocese of St. Paul , 817 N.W.2d 150, 164 (Minn. 2012). When analyzing whether an expert's opinion has foundational reliability, a court must consider the purpose for which the testimony is being offered, the reliability of the underlying theory, and the reliability of the evidence in the particular case. Id. at 167-69.
We turn first to Stewart's challenge to the reliability of the underlying theory. In opposing Dr. Swenson's testimony, Stewart asserted that the theory of abusive *676head trauma is unreliable because it has been "call[ed] into question." He argues that the district court abused its discretion by admitting the testimony because the state failed to counter this argument with evidence of the theory's reliability. This argument is unavailing.
Rule 702 establishes reliability, not universal acceptance or credibility, as the threshold for admissibility. A district court analyzing whether to admit expert opinion testimony has discretion to determine whether the theory underlying the opinion is reliable, consistent, and accurate. Id. at 168. But it may not assess the weight and credibility of the expert's opinion. Such determinations are the province of the fact-finder. Pfeiffer v. Allina Health Sys. , 851 N.W.2d 626, 639 (Minn. App. 2014), review denied (Minn. Oct. 14, 2014); see also State v. Roberts , 876 N.W.2d 863, 868 (Minn. 2016) (recognizing that appellate courts afford the fact-finder "broad deference ... in determining the appropriate weight to assign expert ... testimony"). Thus, when the record indicates that proffered expert opinion testimony is based on a reliable scientific theory, the fact that other theories or evidence tend to contradict the opinion does not justify its exclusion. That is the case here.5
The record amply establishes that the theory of abusive head trauma is reliable. In support of his motion in limine, Stewart himself submitted a Swedish health agency's report that, while criticizing the theory of abusive head trauma, indicated that abusive head trauma is a "conventional[ ]" diagnosis, recognized by the American Academy of Pediatrics. Stewart's proffered report further stated that the diagnosis applies when a child exhibits the "triad" of bleeding around the brain, brain injury, and retinal hemorrhages, although the child's medical history must be considered more broadly to rule out other causes. The state likewise asserted that abusive head trauma is the "accepted" diagnosis when a child presents with such injuries and no history of accidental trauma, and that "the testimony of medical experts on the cause and nature of these injuries is long standing."6 On this record, the district court properly rejected Stewart's argument that the theory of abusive head trauma is unreliable.
We are similarly unpersuaded by Stewart's argument that Dr. Swenson's testimony is unreliable in this particular case because she failed to perform a thorough differential diagnosis. While the state did not submit Dr. Swenson's report in response to Stewart's motion, its written arguments and the complaint reflect the doctor's thorough diagnostic process. Both reveal that Dr. Swenson identified abusive head trauma, inflicted immediately before B.G.D. became symptomatic, as the cause of his injuries. And the state indicated her opinions were not merely based on B.G.D.'s exhibition of the three indicators of abusive head trauma, but also the absence of other viable explanations for those injuries-a severe accidental trauma, B.G.D.'s "head banging," or a coagulation or bleeding disorder.
In sum, the district court did not abuse its discretion by determining that Dr. Swenson's opinion testimony had sufficient *677foundational reliability to be admitted at trial.7
Intent Testimony
An expert witness may not offer an opinion as to a defendant's intent. State v. Xiong , 829 N.W.2d 391, 396 (Minn. 2013). "[A] juror and an expert are equally positioned to make determinations of intent and it is the juror's job to make such determinations, not the expert's as a thirteenth juror." State v. Prtine , 784 N.W.2d 303, 313 (Minn. 2010). But an expert may testify as to an opinion that "embraces" that ultimate issue. Minn. R. Evid. 704. For example, a medical expert may inform the jury's determination as to the defendant's mental state by testifying to the nature and possible cause of a victim's injuries, including whether they "could or could not have been the result of accident." Prtine , 784 N.W.2d at 313.
That is what Dr. Swenson did here. She testified to her opinion-based on B.G.D.'s medical history, the extent and nature of his injuries, his lab results, and her professional understanding of the likelihood of other causes-that B.G.D.'s injuries could not be accidental. She also offered her professional opinion as to the timing of B.G.D.'s injuries, which similarly points to Stewart's intentional actions as the cause of B.G.D.'s injuries. Overall, Dr. Swenson's testimony did not purport to decide the question of Stewart's intent for the jury; it provided medical context from which the jury made its own determination as to whether Stewart intentionally inflicted B.G.D.'s injuries. We discern no abuse of discretion in admitting such testimony.8
III. Stewart's criminal-history score must be corrected.
The Minnesota Sentencing Guidelines provide presumptive sentences based on the severity level of the offense and the offender's criminal-history score. Minn. Sent. Guidelines 2 (Supp. 2015). We review interpretation of the sentencing guidelines de novo, " 'apply[ing] the rules of statutory construction.' " State v. Oreskovich , 915 N.W.2d 920, 926 (Minn. App. 2018) (quoting State v. Campbell , 814 N.W.2d 1, 4 (Minn. 2012) ). If the language of a sentencing guideline is "plain and unambiguous," we will give it effect. Id. Only if a guideline is susceptible of more than one reasonable interpretation will we look further to canons of statutory construction. Id.
A district court determines a defendant's criminal-history score based on his prior convictions or adjudications and custody status at the time of the current offense. Minn. Sent. Guidelines 2.B. A felony conviction is assigned criminal-history points according to the severity level of the prior offense, but only if a "felony sentence" was stayed or imposed before the *678current sentencing. Minn. Sent. Guidelines 2.B.1. The sentencing guidelines establish a different calculation for felony convictions that result in non-felony sentences: "Except when a monetary threshold determines the offense classification of the prior offense (see section 2.B.7), when a prior felony conviction resulted in a non-felony sentence (misdemeanor or gross misdemeanor), the conviction must be counted in the criminal history score as a misdemeanor or gross misdemeanor conviction ...." Minn. Sent. Guidelines 2.B.1.h.
The district court assigned Stewart two criminal-history points based on his October 2014 conviction for theft of property exceeding $5,000 in value: one felony point and one custody-status point (because Stewart was on probation at the time of the current offense). Stewart challenges the felony point, arguing that section 2.B.1.h unambiguously precludes it because the theft conviction resulted in a gross misdemeanor sentence.
We begin our analysis by examining Stewart's claim that he received a gross misdemeanor sentence. The record reflects that Stewart initially received a stay of imposition and was placed on probation. Had he simply completed probation, his conviction would have been deemed a misdemeanor under Minn. Stat. § 609.13, subd. 1(2) (2018). But it would be considered a felony conviction for purposes of his criminal-history score. Minn. Sent. Guidelines 2.B.1; see Campbell , 814 N.W.2d at 7 (holding that the precursor to section 2.B.1.h applies only when "a gross misdemeanor sentence was imposed for the felony conviction," not when the defendant received a stay of imposition). In July 2016, the district court "amended" the sentence and executed a sentence of 342 days.9 Because this sentence is within gross misdemeanor sentencing limits, see Minn. Stat. § 609.02, subd. 4 (2018), Stewart's prior conviction is deemed a gross misdemeanor under Minn. Stat. § 609.13, subd. 1(1) (2018). We therefore agree with Stewart that he received a gross misdemeanor sentence for purposes of section 2.B.1.h.
Despite the gross misdemeanor sentence, the state argues that Stewart's theft conviction is excluded from the reach of 2.B.1.h because it was a felony conviction based on a "monetary threshold." The state acknowledges that excluding all theft offenses likely is not the intended result of section 2.B.1.h, but asserts that courts must give effect to a sentencing guideline's plain language, even if it produces an apparently anomalous result. See State v. Smith , 899 N.W.2d 120, 125 (Minn. 2017).
We agree with the state that the plain language of the monetary-threshold exception appears to make section 2.B.1.h inapplicable to offenses such as Stewart's theft. But Stewart highlights two aspects of section 2.B.1.h that convince us this is not the only reasonable interpretation.
First, Stewart points to the parenthetical instruction to "see section 2.B.7." That section establishes that a prior offense generally is classified as a felony or lesser offense based on the offense definition in effect at the time of the current offense. Minn. Sent. Guidelines 2.B.7. But that section clarifies that "when a monetary *679threshold determines the offense classification," the offense is classified based on "the monetary threshold in effect when the prior offense was committed." Id. Thus, the monetary-threshold exception in section 2.B.1.h, viewed in light of section 2.B.7, does not exclude all monetary-threshold offenses from its reach; the exception simply establishes when a prior conviction for a monetary-threshold offense is considered a "prior felony conviction" to which section 2.B.1.h may apply. We recognize this interpretation is imperfect because it treats the monetary-threshold exception as essentially a superfluous clarification. See State v. Reyes , 890 N.W.2d 406, 410 (Minn. App. 2017) (stating that statutory words or phrases should not be deemed superfluous but recognizing that reasonable interpretations may contain superfluous terms). But it better accounts for all of the language of section 2.B.1.h than the state's interpretation.
Second, Stewart notes that the monetary-threshold exception applies only if the threshold "determines" the classification. This term mirrors the language of section 2.B.7 and recognizes that an offense's monetary threshold is not the only factor that may determine its classification. Rather, classifications are "determined" not only by current offense definitions or the monetary threshold at the time of the prior offense but also by "sentencing policies." Minn. Sent. Guidelines 2.B.7. Such sentencing policies include a district court's authority to stay imposition of a sentence, with the result that a felony conviction is deemed to be for a misdemeanor or gross-misdemeanor offense, Minn. Stat. § 609.13, subd. 1, "but is still included in criminal history under section 2.B." Minn. Sent. Guidelines 1.B.19.a (Supp. 2015).
Based on these considerations, we conclude that the language of section 2.B.1.h reasonably supports both the state's and Stewart's interpretations. But Stewart's interpretation better accounts for all of the section's language and its relation to other guidelines provisions. To the extent ambiguity persists, several additional considerations persuade us that Stewart's interpretation is correct.
First, the comments of the Minnesota Sentencing Guidelines Commission consistently indicate it intended section 2.B.1.h's exception to establish when a prior conviction for a monetary-threshold offense is considered a "prior felony conviction." See Campbell , 814 N.W.2d at 5 (endorsing consideration of "commission policy and official commission interpretation" when interpreting an ambiguous provision of the sentencing guidelines). In addressing criminal-history calculation generally, the commission stated: "When an offender was convicted of a felony but was given a misdemeanor or gross misdemeanor sentence, the offense will be counted as a misdemeanor or gross misdemeanor for purposes of computing the criminal history score." Minn. Sent. Guidelines cmt. 2.B.111. And similarly, in addressing offense classification, the commission expressly "recognized that the classification of criminal conduct as a felony, gross misdemeanor, misdemeanor, or petty misdemeanor is determined legally by the sentence given rather than the conviction offense." Minn. Sent. Guidelines cmt. 2.B.701. In short, the commission plainly expressed its intention that-without exception-the sentence is the controlling factor in determining whether and how to count a prior offense in a criminal-history score.
Second, the history of section 2.B.1.h confirms that the commission did not intend to exclude all monetary-threshold offenses from its reach. From the time the sentencing guidelines were adopted until 2011, the language of section 2.B.1.h contained *680no exception. Compare Minn. Sent. Guidelines II.B.1.c (1980) with Minn. Sent. Guidelines 2.B.1.e (Supp. 2011). And the only references to monetary thresholds were in offense-classification sections of the guidelines. See Minn. Sent. Guidelines 2.B, cmt. 2.B.502 (Supp. 2011). In 2012, the commission undertook a large-scale "primarily stylistic" revision to make the guidelines "easier to read, use, and understand." Minn. Sent. Guidelines Comm'n, Guidelines Revision Project, Adopted Modifications 2012 at 3 (April 2012) http://mn.gov/msgc-stat/documents/Rewrite%20Project/Revision%20Project%20Adopted%20Modifications.pdf. As part of the revision, the commission moved the monetary-threshold language to section 2.B.1.h, creating what now looks like an exception. Id. at 24, 30. The commission did not indicate that this "stylistic" change was intended to break from the long-standing practice of applying the principle of section 2.B.1.h to all convictions.
In sum, the language of section 2.B.1.h, the commission's comments, and the section's historically inclusive application indicate that a felony conviction that results in a gross misdemeanor or misdemeanor sentence should be treated as a gross misdemeanor or misdemeanor for purposes of calculating a defendant's criminal-history score. Because Stewart's felony theft conviction resulted in a gross misdemeanor sentence, he should not have been assigned a second criminal-history point. Accordingly, we reverse Stewart's sentence and remand for the district court to resentence based on the correct criminal-history score of one.
DECISION
We affirm Stewart's conviction because it is supported by sufficient evidence, including properly admitted expert testimony about the cause of B.G.D.'s injuries. But because Stewart's sentence is based on a criminal-history score that incorrectly includes a felony point for a conviction that received a non-felony sentence, we reverse his sentence and remand for resentencing.
Affirmed in part, reversed in part, and remanded.
3.
Great bodily harm is defined as "bodily injury which creates a high probability of death, or which causes serious permanent disfigurement, or which causes a permanent or protracted loss or impairment of the function of any bodily member or organ or other serious bodily harm." Minn. Stat. § 609.02, subd. 8 (2014).
4.
Stewart does not claim that the theory of abusive head trauma is "novel," such that the state must establish that it is "generally accepted in the relevant scientific community." See Minn. R. Evid. 702.
5.
We note this case differs from In re: 3M Bair Hugger Litigation , --- N.W.2d ----, ----, No. A18-0473, slip op. at 11, 2019 WL 178498 (Minn. App. Jan. 14, 2019), in which this court affirmed the exclusion of evidence regarding a novel scientific theory.
6.
Consistent with these pretrial submissions, Dr. Swenson testified (without objection) that the abusive head-trauma diagnosis is not universally accepted, but the American Academy of Pediatrics endorses it and "[h]undreds, probably thousands" of studies support it.
7.
Stewart also contends the district court erred by failing to make express findings addressing the factors at issue in Doe . Because our review of the record supports the evidentiary ruling, the district court's "failure to make a detailed record of its analysis" does not warrant reversal. In re Source Code Evidentiary Hearings , 816 N.W.2d 525, 537 (Minn. 2012).
8.
Stewart also urges us to reverse his conviction because "grave doubts" exist as to whether he committed the assault. The supreme court may exercise its supervisory power to overturn a conviction "in the interests of justice" but generally will not order a new trial absent prejudicial error. State v. Beecroft , 813 N.W.2d 814, 846 (Minn. 2012). We are "an error-correcting court," State v. Grigsby , 806 N.W.2d 101, 114 (Minn. App. 2011), aff'd , 818 N.W.2d 511 (Minn. 2012), and lack the supreme court's supervisory power, see State v. Ramey , 721 N.W.2d 294, 302 n.6 (Minn. 2006). Because we discern no trial error, we will not disturb Stewart's conviction.
9.
The state asserts that the record "does not clearly reveal what actually occurred with [Stewart's] sentence." We disagree. The presentence investigation report and Stewart's sentencing argument both reflect a stay of imposition followed by an executed 342-day sentence, and the state, which bore the burden of proof, presented no contrary evidence. See Williams v. State , 910 N.W.2d 736, 740 (Minn. 2018) ("The State bears the burden of proof at sentencing to show that a prior conviction qualifies for inclusion within the criminal-history score.").