ROVNER, Circuit Judge.

Dr. Jaswinder Rai Chhibber was charged with eight counts of making false statements relating to health care matters, in violation of 18 U.S.C. § 1035, and eight counts of health care fraud, in violation of 18 U.S.C. § 1347. A jury found him guilty of four counts of making false statements and five counts of health care fraud. He appeals, challenging several evidentiary decisions made by the district court. We affirm.

I.

Chhibber was an internist who operated the Cottage Grove Community Medical Clinic (hereafter the "Clinic"), a walk-in medical office on the south side of Chicago. For patients who had insurance or were covered by Medicare, Chhibber ordered an unusually large volume of diagnostic tests, including echocardiograms, electrocardiograms, pulmonary function tests, nerve conduction studies, carotid Doppler ultrasound scans and abdominal ultrasound scans. Chhibber owned the equipment to perform these tests and his staff performed them at the Clinic according to his orders. The government asserted that Chhibber obtained reimbursements for the tests from insurers by presenting claims that contained false and misleading diagnostic codes to justify the tests. He also supported the claims by recording fake symptoms and sham diagnoses in his patients' medical charts.

At trial, the government presented witnesses who had worked for Chhibber, patients who saw him, and undercover agents who presented themselves to the Clinic as persons needing medical services. This testimony revealed that Chhibber was in the habit of ordering diagnostic tests for his patients regardless of any symptoms they exhibited or reported. Chhibber's former employees testified that he often ordered tests before he even arrived at the office, based on phone calls with staff in which he inquired about little more than the names of the patients' insurers and when the patients had last been given tests. The employees performed the tests themselves with little training, and the results were not reviewed by specialists. Indeed, Chhibber refused to provide formal training, citing cost as a reason; instead he required employees to train each other. Training for some employees lasted only minutes. To the extent that the tests were reviewed by anyone, Chhibber himself performed the review even though he was not board certified in any medical specialty or subspecialty. In usual practice, though, the tests were not reviewed at all. Chhibber's patients and the undercover officers who posed as patients testified that they did not report the symptoms that Chhibber recorded in their charts and that Chhibber did not discuss with them the results of tests or the serious medical conditions that he attributed to them.

The government also presented testimony from an internist, Dr. Daniel Herdeman, and from two statisticians.1 The government initially sought to qualify the statisticians as experts, and sought to introduce charts demonstrating that Chhibber performed various tests on his patients with much greater frequency than other internists in the same geographical area. Chhibber objected to qualifying the statisticians as experts and also objected to the admission of charts comparing the frequency of Chhibber's testing to the frequency of testing by his peers. He contended that the statisticians had used a peer group that was not truly comparable because they had not considered whether the physicians in that group owned testing equipment as Chhibber did. Chhibber argued that because an unknown number of those doctors in the peer group could simply be referring patients out for the same amount of testing at other facilities, the comparison was neither accurate nor fair. The district court agreed and declined to qualify the statisticians as experts, and also declined to allow the government to present the charts comparing Chhibber's frequency of performing tests to the frequency of other doctors.

But the court did allow the statisticians to present summary charts containing only the percentages of Chhibber's patients who received certain tests or diagnoses, without any comparison to a peer group. Exhibit 801X portrayed the percentage of Chhibber's patients who were insured by Blue Cross Blue Shield of Illinois and who received (1) a carotid Doppler (60.47%); (2) a transcranial Doppler (38.74%); (3) an electrocardiogram (34.55%); (4) an echocardiogram (55.50%); (5) a nerve conduction study (18.32%); (6) a pulmonary function test (59.69%); and (7) an abdominal ultrasound (46.07%). A corresponding chart, Exhibit 803X, showed the percentage of Chhibber's Blue Cross patients receiving particular diagnoses: (1) cardiac murmurs (63.09%); (2) shortness of breath and chest pain (62.04%); and (3) hypertension (33.51%). Similar charts were entered into evidence for patients of Chhibber who were covered by Medicare. Exhibit 806X revealed that large numbers of Chhibber's Medicare patients received (1) a carotid Doppler (64.79%); (2) a transcranial Doppler (30.75%); (3) an electrocardiogram (48.59%); (4) an echocardiogram (62.68%); (5) a nerve conduction study (21.13%); and (6) a pulmonary function test (73.24%). Exhibit 808X portrayed the percentage of Chhibber's Medicare patients receiving particular diagnoses: (1) cardiac murmurs (62.21%); (2) shortness of breath and chest pain (66.20%) and (3) hypertension (50.70%).

Chhibber objected to the introduction of these four exhibits on the grounds that they were prejudicial and irrelevant. He protested that it was improper for the government to argue that a high percentage for a particular procedure or diagnosis implied fraud unless someone testified that the number was comparatively high, and that the jury could not know what the raw numbers meant without an appropriate comparison. He also contended that the percentages distorted reality because they were calculated on a per patient basis rather than a per visit basis. Such an assessment did not account for occasions where a single patient came in for twenty or thirty visits and received only one test. The court overruled the objection and allowed the charts to come in under Federal Rule of Evidence 1006, allowing the statisticians to testify as summarizers of voluminous records. The jury, as we noted above, convicted Chhibber on some of the counts and acquitted him on others. He appeals.

II.

On appeal, Chhibber contends that the trial court improperly admitted the four exhibits described above. He also argues that the government should have been required, as a matter of law, to present expert testimony that the treatment Chhibber provided was medically unnecessary. Finally, he contends that the cumulative effect of the trial court's errors deprived him of the right to a fair trial.

A.

We review the court's decision to admit or exclude evidence for abuse of discretion. United States v. Simon, 727 F.3d 682, 696 (7th Cir.2013); United States v. Thornton, 642 F.3d 599, 604 (7th Cir. 2011). See also United States v. Isaacs, 593 F.3d 517, 527 (7th Cir.2010) (reviewing a district court's decision to admit summary charts into evidence under Rule 1006 for abuse of discretion). Chhibber contends that, once the district court properly excluded the proposed peer group evidence, the remaining charts showing how often Chhibber performed certain tests and made particular diagnoses were irrelevant and prejudicial. Without a point of comparison, Chhibber maintains that the jury had no basis for inferring fraud from these numbers. He also complains that the numbers were misleading because they were calculated on a per-patient basis rather than a per-visit basis. The numbers were prejudicial, he claims, because the government argued that his testing numbers were so high that they reflected an "impossible volume" without giving the jury anything to compare them to. Finally, he contends that the statistics were the sole basis for any claim of Medicare fraud against him because no Medicare patients testified at trial. The numbers alone, he maintains, must have been prejudicial because there was no other basis on which to convict him of Medicare fraud.

The district court did not abuse its discretion in admitting these four summary charts. First, we note that Chhibber does not contend that the numbers were anything other than an accurate summary of his billing and medical records. That is, the charts accurately portrayed the data culled from his records, and were not misrepresented to the jury as anything other than what they actually were. His objection that the numbers were calculated on a per-patient rather than a pervisit basis, for example, was a point that he could argue to the jury, and he was free to argue that the numbers lacked significance for this reason. But the numbers were not portrayed to the jury as per-visit when in fact they were per-patient; all of the percentages were accurate summaries of Chhibber's own records. See Isaacs, 593 F.3d at 527-28 (summary exhibits which accurately portray the underlying data are admissible so long as the proponent complies with the dictates of Rule 1006).

More importantly, there was evidence in the record that provided a point of comparison for the jury. Although the government's peer group statistics were excluded because the court found that the government's proposed peer group was not comparable, the government presented the testimony of Dr. Herdeman, an internist who was qualified as an expert in the field of internal medicine and diagnosis. R. 249, Tr. at 384. Chhibber reserved his right to object to Dr. Herdeman's expertise "as to the giving of tests," but subsequently he did not object to extensive testimony by Dr. Herdeman regarding how often an internist encounters patients with the conditions diagnosed by Chhibber and how often an internist typically orders the tests performed by Chhibber.

Dr. Herdeman's testimony provided an ample basis for the jury to evaluate the statistical summaries. Dr. Herdeman is a board certified internist who had been practicing medicine in a variety of settings for twenty-nine years at the time of the trial. He had worked for approximately fourteen years at the Swedish American Medical Group in Rockford, Illinois, serving a population that included both middle class and impoverished patients of all races and adult age groups. Prior to that, he worked at a community health clinic where he was both an internist and medical director, supervising other internists, family doctors and pediatricians. The community health clinic served a patient population that was at or below the poverty level, again spanning all racial groups and age ranges. Prior jobs provided him with similar experience, practicing internal medicine, and supervising other internists. In addition, for approximately twenty years, he was a clinical instructor at the University of Illinois medical branch in Rockford, Illinois. Dr. Herdeman's background gave him extensive experience with patients similar to those treated by Chhibber in the Clinic.

Dr. Herdeman testified that echocardiograms, electrocardiograms, carotid Doppler scans, transcranial Doppler scans, pulmonary function tests, and nerve conduction studies are not routine screening procedures but are specialized tests performed by persons with special training. Dr. Herdeman stated that a doctor would not order any of these specialized tests without examining a patient because only certain conditions would justify performing each test. Based on his training and experience, he expected that an internist would order (1) one or two echocardiograms per month; (2) approximately one or two electrocardiograms per week, or approximately 2% to 5% of his total patient population; (3) ten to fifteen carotid Doppler scans per year; (4) ten to twelve pulmonary function tests per year; and (5) five to eight nerve conduction studies per year. Dr. Herdeman had never ordered a transcranial Doppler scan because none of his patients had ever needed one. In his experience in working with older patient populations and more impoverished patients, he testified that the numbers were very similar, with only marginal increases. Dr. Herdeman also explained that echocardiograms, carotid Doppler scans, transcranial Doppler scans, pulmonary function tests, and nerve conduction studies were not typically performed in an internist's office because most internists are not trained to perform the test, and the small number of tests needed would not justify hiring a trained technician. In his experience, patients needing these tests were referred to hospitals that had the personnel and equipment necessary to perform the tests. Dr. Herdeman also explained that he did not feel qualified as an internist to interpret the images generated by an echocardiogram, carotid Doppler scans, transcranial Doppler scans, pulmonary function tests, and nerve conduction studies.2 Typically, cardiologists review the raw data from echocardiograms, radiologists interpret the data from Doppler scans, pulmonologists review pulmonary function test data, and neurologists review the results of nerve conduction studies. For each type of test, the appropriate specialist generates a report for the patients' internists.

The government also questioned Dr. Herdeman about Chhibber's patient progress notes for the patients where fraud was alleged. Dr. Herdeman testified that the notes did not contain the information that he would expect to find for persons given the diagnoses that Chhibber assigned to these patients. For example, for a patient with a heart murmur, Dr. Herdeman expected to see answers to a number of questions about the patient's symptoms during various activities, and details about the sound of the murmur when the patient was in a variety of positions. He would expect to see questions about prior events and diagnoses, medical history, and an indication that the condition was discussed with the patient. For a patient experiencing chest pain, he would expect to see answers to numerous follow-up questions to determine if the chest pain was caused by a benign condition or a more serious cardiac issue. Similarly for a patient with carotid bruit, an unusual sound in the carotid arteries indicating a possible dangerous blockage, Dr. Herdeman would expect a number of follow-up questions to be answered in the patient's chart. In the case of one twenty-eight year old patient where Chhibber diagnosed carotid bruit, Dr. Herdeman noted that this was extremely unusual and worrisome in a patient this age, that he would ask certain questions and record the answers, and that he would discuss this condition with the patient because of the potentially serious health consequences of the conditions it might indicate. In another patient that Chhibber noted as suffering from abdominal pain, Dr. Herdeman remarked that he would expect to see (and did not see on the patient's chart) answers to a number of follow-up questions to determine the nature and seriousness of the pain. For at least four patients, the charts indicated in the results of an echocardiogram the absence of "intracranial shunts." Given that the tests were ultrasounds of the heart, a finding of no shunts in the cranium (the head) of the patient was a nonsensical notation. Each of the charts was missing critical information about the potentially serious symptoms that the patients purportedly experienced.

Contrary to Chhibber's claim, Dr. Herdeman provided an adequate basis for comparison and did not function as a "peer group of one." He was an expert, who was testifying about his experiences both as an internist and a supervisor of other internists for more than twenty-nine years. Chhibber did not object to any of Dr. Herdeman's testimony regarding the frequency with which an internist would order these tests. And as should now be obvious, Chhibber ordered the tests for his patients at considerably higher rates than internists generally do, and he did so without evaluating the patients for need, with staff that was poorly trained to administer the tests, and without the expertise to properly evaluate the results. This evidence was relevant to corroborate the claims of Chhibber's actual patients and the undercover officer-patients that they did not report (or display) to Chhibber the symptoms he recorded on their charts to justify the tests he ordered. The evidence also demonstrated that the tests normally would not have been ordered without thorough examinations of the patients, and that the errors in Chhibber's notes were not incidental or typographical but were part of a pervasive fraud scheme. This evidence also corroborated the testimony of Chhibber's former employees who testified that Chhibber tested patients in large numbers without first examining them and without ever interpreting the results of the tests. In short, the evidence was relevant to proving the government's case and was not presented in the vacuum that Chhibber suggests. Instead, the charts were presented in the context of Dr. Herdeman's expert testimony, as well as in the framework of the testimony of patients (both real and undercover) and former employees who attested to their experiences with Chhibber's practices. There was no abuse of discretion in admitting the challenged summary charts.

B.

Chhibber next argues that the government should have been required to present expert testimony to prove that the tests he provided were not medically necessary and that his diagnoses were incorrect. He asserts that the question of whether expert testimony is required to sustain a conviction for the provision of medically unnecessary care under 18 U.S.C. §§ 1347 and 1035 is one of first impression. We will overturn a verdict for insufficiency of the evidence only if, after viewing the evidence in the light most favorable to the government, the record is devoid of evidence from which a rational trier of fact could find guilt beyond a reasonable doubt. United States v. McIntosh, 702 F.3d 381, 385 (7th Cir.2012), cert. denied, ___ U.S. ___, 133 S.Ct. 1484, 185 L.Ed.2d 383 (2013); United States v. Vaughn, 585 F.3d 1024, 1028 (7th Cir. 2009), cert. denied, 560 U.S. 954, 130 S.Ct. 3385, 177 L.Ed.2d 306 (2010); United States v. Olson, 450 F.3d 655, 664 (7th Cir.2006).

Chhibber was convicted of four counts of false statements under section 1035, and five counts of health care fraud under section 1347. Section 1035 provides, in relevant part:

18 U.S.C. § 1035(a)(2). Section 1347 provides, in relevant part:

18 U.S.C. § 1347(a).

Although neither statute refers expressly to the provision of medically unnecessary treatment, the indictment charged that Chhibber violated these statutes by ordering medically unnecessary tests for patients covered by health care benefit programs and then recording false diagnosis codes in those patients' charts and on reimbursement claim forms in order to justify giving the tests. The indictment further alleged that, as part of this scheme, Chhibber ordered tests without first examining patients, that he sought reimbursement for the tests even when he failed to review the results, that he ordered tests on patients he had examined but whose signs and symptoms did not indicate that the tests were medically necessary, that he documented false and fictitious signs and symptoms in patients' medical charts to provide written support for the tests ordered, that he and his staff (at his direction) added notes to patients' charts weeks or months after tests had been performed to make it appear as if he had reviewed the results of the tests when he had not in fact done so, and that he caused insurance claims to be filed for medically unnecessary tests. The individual counts of the indictment charged Chhibber with submitting claims to Blue Cross Blue Shield for tests performed on seven different patients3 when those claims falsely indicated that the patients suffered from particular medical symptoms (in violation of section 1347), and with noting false diagnoses in the medical charts of seven patients (in violation of section 1035).

We need not decide as a matter of law whether expert testimony is required on the issue of medical necessity because the government did in fact provide expert testimony from Dr. Herdeman regarding what conditions justified particular tests. Dr. Herdeman testified that each test at issue was not a general screening tool but was used only when patients exhibited particular signs and symptoms that required further investigation. Absent those particular signs and symptoms, Dr. Herdeman stated, the tests were not medically necessary. Dr. Herdeman also testified that he would never order any of these tests without first examining a patient to determine the need for testing. See United States v. Hunt, 521 F.3d 636, 645 (6th Cir.2008) (evidence adequate to prove health care fraud under section 1347 where claims were submitted to insurer for tests that were ordered without in-person examination of patient, and where government's expert testified that an examination of the patient would be required to determine medical necessity of the tests); United States v. Morgan, 505 F.3d 332, 341-42 (5th Cir.2007) (a reasonable jury could conclude that physician violated section 1347 when she signed prescriptions for durable medical equipment that was billed to Medicare without ever examining the patients). Dr. Herdeman explained what the medical records would look like in the presence of a legitimate examination for the serious symptoms and conditions that Chhibber recorded on his patients' charts, and he testified that the test results would be reviewed by specialists and discussed with the patients, none of which happened in any of the charged counts. The government also provided corporate representatives from Blue Cross and Medicare who defined medical necessity and described how claims would be paid. The corporate representatives verified that they would not pay for claims founded on fictitious or fraudulent diagnoses.

And there was ample evidence from the patients and undercover officers that the patients did not report the symptoms recorded by Chhibber or that he ordered tests without examining the patients at all. For example, for the section 1347 counts, Chhibber submitted claims to Blue Cross for (1) an electrocardiogram and pulmonary function test for an undercover agent which Chhibber justified by claiming that the agent suffered shortness of breath, a symptom that the agent neither reported nor displayed; (2) an echocardiogram and a transcranial Doppler scan for another undercover agent which Chhibber justified by claiming that the agent had heart murmurs, syncope and collapse, when the agent had never reported syncope and collapse and when Chhibber never mentioned the alleged heart murmur to the agent and never noted any proper investigation in the chart; and (3) an electrocardiogram, a pulmonary function test, an echocardiogram and a carotid Doppler for a twenty-eight year-old patient who came to Chhibber's office to obtain a nursing school registration physical, tests Chhibber justified by claiming the woman had complained of chest pain, shortness of breath and dizziness, symptoms she never reported. In addition, Chhibber diagnosed this patient with a heart murmur and a carotid bruit, both very serious conditions (especially for a twenty-eight year-old patient) and yet never mentioned these diagnoses to the patient and never told her the results of any of these tests. In combination with Dr. Herdeman's testimony and that of the corporate representatives, this evidence was more than adequate to support the conviction on the section 1347 counts.

As for the section 1035 claims, Chhibber documented these made-up medical symptoms and diagnoses in his patients' charts in order to backstop his claims for benefits from Blue Cross Blue Shield. United States v. Natale, 719 F.3d 719, 742 (7th Cir.2013), petition for cert. filed, ___ U.S.L.W. ___ (Dec. 20, 2013) (No. 13-744) (setting forth the elements that must be proved for a conviction for false statements relating to health care matters under section 1035(a)(2)). For the agents mentioned above, Chhibber recorded shortness of breath when the agent was not suffering from and had not reported shortness of breath. Chhibber recorded syncope (fainting) and collapse for a different agent who was not suffering from and had not reported these symptoms. For the twenty-eight year-old woman, he recorded chest pain, dizziness and shortness of breath in her chart when she neither reported nor exhibited these conditions. The evidence was similar on the other counts of conviction. Deciding whether to credit simple factual testimony from these witnesses, testimony that conflicted with Chhibber's written records, required no expert testimony and was well within the province of the jury. On appeal, we do not reweigh the evidence or assess witness credibility. United States v. Wasson, 679 F.3d 938, 949 (7th Cir.2012), cert. denied, ___ U.S. ___, 133 S.Ct. 1581, 185 L.Ed.2d 576 (2013). See also United States v. Carraway, 612 F.3d 642, 645 (7th Cir.2010), cert. denied, ___ U.S. ___, 131 S.Ct. 1025, 178 L.Ed.2d 848 (2011) (credibility determinations are best handled by the trier of fact, not the appellate court). This evidence was more than sufficient to sustain the convictions on the section 1035 counts.

III.

Given that there were no errors in the district court's rulings, we need not consider Chhibber's claim of cumulative error. We have reviewed his other claims and find no merit in them. In sum, the court did not err in admitting charts that accurately summarized voluminous records, records that were relevant to the government's case. We need not decide whether expert testimony was necessary to prove the question of medical necessity because the government did provide an expert who explained the symptoms, diagnoses, and investigations that would justify the tests that Chhibber ordered. In light of the wealth of evidence that Chhibber sought reimbursement for tests he ordered in the absence of any justification and the evidence that he fabricated patient charts to justify his actions, the judgment of the district court is AFFIRMED.