(dissenting, with whom Kantrowitz, J., joins). Driven by the critical shortcomings of the plaintiff’s case, which I conclude fall woefully short of the standard demanded by G. L. c. 231, § 60B, I respectfully dissent. The vital question here, as to which the parties sharply disagree, is whether the plaintiff’s proof permits an inference that the defendant did not, in fact,1 conform to good medical practice. See Blood v. Lea, 403 Mass. 430, 433 (1988); Booth v. Silva, 36 Mass. App. Ct. 16, 20 (1994). As is true in most instances, this inquiry can be answered only with the aid of expert opinion. Kapp v. Ballantine, 380 Mass. 186, 190 & n.4 (1980). The expert’s opinion, however, must be rooted in the record evidence and not be based on speculation, conjecture, or assumptions not supported by the evidence. Blood v. Lea, supra at 434.
When comparing Dr. Kenneth C. Fischer’s expert opinion as to what would have been good medical practice against the record evidence memorializing the actual treatment rendered by the defendant, Dr. Hilarie Cranmer, a remarkable convergence emerges between the expert opinion and the treatment provided to the plaintiff on March 22,2008, in the emergency department at Brigham and Women’s Hospital (BWH). Promptly, upon interviewing and examining the plaintiff, Dr. Cranmer recognized and appreciated the plaintiff’s symptoms and history as involving a hypertensive crisis and transient ischemic attack (TIA), as is demonstrated by her orders to administer antihypertensive agents to lower the plaintiff’s blood pressure; perform a computer tomography (CT) scan of the plaintiff’s head; conduct laboratory work *683of the plaintiff’s blood and fluids; and schedule magnetic resonance imaging/angiography (MRI/A) scans to evaluate the plaintiff for a head stroke.
As of 12:45 p.m., while under the care of Dr. Cranmer and assisting medical personnel, the plaintiff’s blood pressure had stabilized within normal limits. Neither the laboratory tests nor CT scan revealed or suggested an acute end-organ dysfunction. Dr. Fischer’s opinion and ultimate conclusion — i.e., that Dr. Cranmer failed to conform to good medical practice — hinges entirely upon the presence of acute end-organ dysfunction or ongoing damage. On this record, there is no evidence at all to permit an inference that the plaintiff suffered, on March 22 at the BWH emergency department, from acute end-organ damage or dysfunction.
It is undisputed that Dr. Cranmer assigned the plaintiff to the emergency department observation section (OBS) for monitoring of a possible TLA and continued blood pressure control. Dr. Cranmer also scheduled MRI/A scans for the latter part of the afternoon. It is also undisputed that, due to the plaintiff’s unwillingness to undergo the MRI/A scans, Dr. Cranmer did not have the benefit of this essential medical diagnostic resource.2
Dr. Fischer explicitly defined a “hypertensive emergency” as “a severe and persistent elevation in the blood pressure with acute impairment of an organ system (end-organ dysfunction).” “[A]rm and leg weakness, paresthesia, gait disturbance, blurry vision, and difficulty with word finding” may suggest dysfunction, but do not demonstrate the presence of actual acute impairment of an end-organ system. The laboratory work, imaging studies, and electrocardiogram (EKG) results here furnished no indication of acute end-organ damage. Under no view of the evidence was it demonstrated that the plaintiff, while treated at the BWH emergency department on March 22, presented a “hypertensive emergency” as that medical term was defined by Dr. Fischer. Rather, it cannot be reasonably disputed that on March 22, the plaintiff presented a nonemergent hypertensive condition, for which Dr. Cranmer pre*684scribed, and the plaintiff received, oral antihypertensive medication, and gradually, over a two-hour period, the plaintiff’s blood pressure was lowered to the normal range. Dr. Cranmer then arranged for the plaintiff to be monitored in the OBS for a TIA and blood pressure control. Dr. Cranmer also requested MRI/A scans to evaluate the plaintiff for a stroke.
The plaintiff refused to undergo the MRI/A scans, and thus we can only speculate what the MRI/A scans would have revealed. Moreover, the next day’s MRI/A scan results shed no light on what they might have revealed the day before.
In short, up to the time of the plaintiff’s discharge on March 22, there were no “indicators” that should have impelled Dr. Cranmer to act other than she did,3 LaFond v. Casey, 43 Mass. App. Ct. 233, 234 (1997); rather, Dr. Fischer’s opinion (and by extension the plaintiff’s malpractice claim) rests on an ill-based factual assumption, namely, the presence of a hypertensive emergency, which, in the end, is not supported by anything in the record. Given this material shortcoming in the proof, combined with the undeniable fact that the plaintiff was unwilling to have the MRI/A scans on March 22, the plaintiff’s proof is legally insufficient to permit an inference in her favor. An inference must be based on “probabilities” not possibilities. Alholm v. Wareham, 371 Mass. 621, 627 (1976). Nor do we look back on this most unfortunate incident “with the wisdom born of the event.” Greene v. Sibley, Lindsay & Curr Co., 257 N.Y. 190, 192 (1931) (Cardozo, C.J.).
Based on the foregoing, I would affirm the judgment of the Superior Court.
Ante at 675.
“Because the standard of care is based on the care that the average qualified physician would provide in similar circumstances, the actions that a particular physician, no matter how skilled, would have taken are not determinative.” Palandjian v. Foster, 446 Mass. 100, 104-105 (2006). “It is permissible to consider the medical resources available to the physician as one circumstance in determining the skill and care required.” Brune v. Belinkoff, 354 Mass. 102, 109 (1968).
The plaintiff’s blood pressure reduced to normal, elevated slightly, and then stabilized at 174/105 — reading the same at 4:00 p.m. and 6:00 p.m., when she was discharged.